| ftp://cs.utk.edu/pub/spike/fig2.ps, *19911002* ~ 3 C ~ 2 C ~ 1 C ~ 3 B ~ 2 B ~ 1 B ~ 4 A ~ 3 A ~ 2 A ~ 1 A |

| ftp://cs.utk.edu/pub/spike/fig3.ps, *19911002* 13 x 12 x 9x 8x 5x4x 4g 3g 2g 1g 4u 3u 2u 1u 3 2 1 Y Y Y W W W 3 2 1 I I I I 4 3 2 1 Cluster |

| ftp://cs.utk.edu/pub/spike/fig1.ps, *19911002* nA n-1 B n-1 C n-1 A n-2 B 3C 3A 2B 2C 2A 1B 1C 1A |

| ftp://cs.utk.edu/pub/pozo/papers/thesis.ps.Z, *19920122* 178 BIBLIOGRAPHY Tankqueray, D., Snelling, D. F., A Distriubted Memory Implementation of the Shallow Water Equations," Dawn of Massively Parallel Processing in Meterology, eds. Hoffman, G.R., Maretis, D. K., Springer-Verlag, 1990. Thinking Machines Corporation, Communications Architecture in |

| ftp://cs.utk.edu/pub/key/co2.ps, *19920123* Volumes of CO2 in Beer Pressure (PSI) Temp 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 30F 1.82 1.92 2.03 2.14 2.23 2.36 2.48 2.60 2.70 2.82 2.93 3.02 31F 1.78 1.88 2.00 2.10 2.20 2.31 2.42 2.54 2.65 2.76 2.86 2.96 32F 1.75 1.85 1.95 2.05 2.16 2.27 2.38 2.48 2.59 |

| ftp://cs.utk.edu/pub/pozo/papers/conpar.ps.Z, *19920614* ularly for the distributed networks, are meant only as a rough guide. Our heuristics have been partially validated by independent static scheduling packages , but further work is needed to compare simluation results with actual multifrontal timings on exisiting architectures. Although we have |

| ftp://cs.utk.edu/pub/pozo/papers/dmf.ps.Z, *19920614* ularly for the distributed networks, are meant only as a rough guide. Our heuristics have been partially validated by independent static scheduling packages , but further work is needed to compare simluation results with actual multifrontal timings on exisiting architectures. Although we have |

| ftp://cs.utk.edu/pub/pozo/papers/scalapack.ps.Z, *19920624* J. Dongarra, J. DuCroz, A. Greenbaum, S. Hammarling, A. McKenney, and D. Sorensen. Lapack: A portable linear algebra library for highperformance computers. In Proceedings of Supercomputing '90, pages 10. IEEE Press, 1990. E. Anderson, Z. Bai, J. Demmel, J. Dongarra, J. DuCroz, A. Greenbaum, S. |

| ftp://cs.utk.edu/pub/pozo/papers/smc.ps.Z, *19920715* A st act We present performance models of numerical computations on the Connection Machine CM-2. Our research identifies machine characteristics that are useful in predicting performance of numerical computations such as matrix/vector operations, relaxation techniques, and PDE applications on this |

| ftp://cs.utk.edu/pub/pozo/papers/smp.ps.Z, *19920923* tr - s t r c i r r r i t r s st s ol an ozo e art ent o o uter c ence n vers t o ennessee s ntroduction We present the design of an object-oriented interface, SMP ( tream essage assing), to program heterogeneous systems. In contrast to other object-oriented solutions for homogeneous systems , , |

| ftp://cs.utk.edu/pub/pozo/papers/siam6th.ps.Z, *19930109* Chapter 1 Performance evaluation of the parallel multifrontal method in a distributed memory environment Roldan Pozo Sharon L. Smith y |

| ftp://cs.utk.edu/pub/plank/papers/USENIX-93W.ps.Z, *19930203* Jgraph A Filter for Plotting Graphs in PostScript James S. Plank Princeton University |

| ftp://cs.utk.edu/pub/chassismib/petewilson-chassis-pres.ps, *19930420* 3Com Proprietary and Confidential Alternative Chassis MIB n Peter Wilson (3Com) 3Com Proprietary and Confidential B a c k C a r d Extension Backplane Main Backplane Fan Tray 1 P S U 1 P S U 2 F r o n t C a |

| ftp://cs.utk.edu/pub/larose/ut-cs-93-195.ps.Z, *19930712* THE DEVELOPMENT AND IMPLEMENTATION OF A PERFORMANCE DATABASE SERVER Brian Howard LaRose Computer Science Department CS-93-195 August 1993 The evelopment and Implementation of a erformance atabase Server A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Brian |

| ftp://cs.utk.edu/pub/larose/short-PDS-paper.ps.Z, *19930712* . 1 . 1 ese rc s r e er r er - - , er r er - - - 1 bstract e r cess er , rc , s r c er e c r s c ers e s s er r e c er sers e rs e c r. s r , ere s c - e ce r e s r er rce r r es c es r ers c ers s erc ers. e rese er e - ccess e er r ce se ser er c c e se e r c c rre e c r er re. s e e s e - sse ser er |

| ftp://cs.utk.edu/pub/larose/pdsug.ps.Z, *19930713* Draft 17 PDS version 1.0 coordinator Draft 16 The following question was posted by Alkiviadis Vazacopoulos at Carnegie Mellon, Pittsburgh, PA: "I would like to obtain the latest version of the paper of Jack J. Dongarra : Performance of Various Computers using Standard Linear |

| ftp://cs.utk.edu/pub/maclennan/CSA-tr.ps.Z, *19930721* Continuous Spatial Automata B. J. MacLennan Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 maclennan@cs.utk.edu CS-90-121 November 26, 1990 |

| ftp://cs.utk.edu/pub/pozo/papers/shpcc94.ps.Z, *19930930* Limited Resource Scheduling in Sparse Matrix Algorithms Roldan Pozo Sharon L. Smith Department of Computer Science Computer Sciences Division University of Tennessee University of California Knoxville, TN 37996-1301 Berkeley, CA. 94270 |

| ftp://cs.utk.edu/pub/key/Skyline-map.ps, *19931001* US-25 US-11 US-70 Delrose Dr. N 523-9060 5103 Skyline Boyd's Bridge Rd Hyatt Riverside Bridge Boyd's Bridge South Knox River Holston Hills Dr. 4 way stop Holston Stop light Magnolia Business Loop I-40 Skyline Chilhowee Asheville Hwy I-640 I-40 Rutledge Pk Cherry St. |

| ftp://cs.utk.edu/pub/pozo/papers/parco93.ps.Z, *19931102* machines over a larger network. It is difficult to achieve true supercomputer performance from a workstation cluster unless the problem is embarrassingly parallel. For example, compare the results shown in Figure 5 from timings on a cluster of IBM RS6000s to the results shown in Figure 6 from timings on |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-121.ps.Z, *19931110* Continuous Spatial Automata B. J. MacLennan Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 maclennan@cs.utk.edu CS-90-121 November 26, 1990 |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/13-FutWk.ps.Z, *19931115* 169 Chapter 10. Future Work A project like CODE can continue nearly forever. There are endless possibilities for enhancements, improvements, and experiments. The basic idea of a model of parallel computation that combines textual specification of sequential computations and graphical composition of them |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/05-OvrVw.ps.Z, *19931115* 10 Chapter 2. Overview of the CODE Language This chapter will informally define the CODE model of parallel computation and provide an overview of how users prepare programs using the CODE environment. The model will be more thoroughly defined in Chapter 3. CODE is a graphical parallel programming |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/17-Ref.ps.Z, *19931115* 187 References W.B. Ackerman, "Data Flow Languages", Computer, Vol. 15. pp.15-25, Feb., 1982. D.A. Adams, A Model for Parallel Compilations , Parallel Processor Systems, Technologies and Applications, pp. 311-334, Spartan/MacMillan, New York, 1968. A.V. Aho, R. Sethi, and J.D. |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/09-Ease.ps.Z, *19931115* 75 Chapter 6. Factors Affecting Ease of Use Ease of use is one of CODE s more important goals since it is one of the central problems of parallel programming. This chapter discusses why parallel programming is more dif ficult than sequential programming, how CODE approaches the goal of ease of use, and |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/10-Impl.ps.Z, *19931115* 92 Chapter 7. Implementation of CODE This chapter describes the implementation of CODE, some of the special problems it addresses, and the techniques used to overcome them. It also mentions some of the tools with which CODE is built. Whenever possible abstract programming tools and techniques have been |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/06-moddef.ps.Z, *19931115* 21 Chapter 3. CODE Model Definition This chapter presents the CODE model of computation at a level of detail sufficient for understanding the example programs in this dissertation. Many of the details of the language will be skipped. See for a complete description of the CODE |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/11-Proto.ps.Z, *19931115* 152 Chapter 8. CODE 2.0 Prototype This chapter briefly discusses a prototype implementation of the CODE 2.0 model of parallel computation. This subject is relevant because it demonstrates that the CODE model can be implemented on a target architecture (Ada ) that is based upon synchronous |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/02-LOF.ps.Z, *19931115* x List of Figures Figure 1-1. CODE Primitive Node. 4 Figure 2-1. A CODE Program. 11 Figure 2-2. UC Node Attribute Form. 12 Figure 2-3. A Dynamic Computation Structure. 17 Figure 2-4. Another Integration Program. 18 Figure 2-5. All CODE Node Types. 19 Figure 3-1. An Example UC Node. 22 Figure 3-2. |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/18-Vita.ps.Z, *19931115* Vita Peter William Newton was born in Omaha, Nebraska on December 26, 1960, the son of John M. Newton and Ann Newton. After graduating from Central High School in Omaha in 1979, he enrolled at the University of Michigan at Ann Arbor where he received the degree of Bachelor of Science in 1982. His major |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/12-RelWk.ps.Z, *19931115* 160 Chapter 9. Related Work Graphs in some form are ubiquitous in models of parallel computation as well as in implementation strategies for parallel languages. However, there are relatively few programming environments in which users directly create parallel programs by the composition of tasks into |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/07-Sem.ps.Z, *19931115* 45 Chapter 4. Computations in the CODE Model This chapter will define the computations that a CODE program can legally perform. It is useful to programmers who wish to create new translation modules for CODE since it defines what is required of the implementation. It is also useful to CODE users since |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/15-ModDef.ps.Z, *19931115* 178 Appendix A. The CODE Model The CODE model is represented as an abstract syntax tree, the nodes of which are C++ objects of a class derived from a special tree class (cAbsTree). The class definitions in concise form follow. Method signatures and other im- plementation specific attributes have been |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/01-TOC.ps.Z, *19931115* vii Table of Contents List of Figures x List of Tables xiii Chapter 1. Introduction 1 1.1 The CODE Approach 1 1.2 Other Approaches 8 Chapter 2. Overview of the CODE Language 10 2.1 A Simple Example Program 10 2.2 CODE Elements Are Actually Templates 17 2.3 Other CODE Nodes 19 2.4 Translating and Running |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/00-FM.ps.Z, *19931115* A GRAPHICAL RETARGETABLE PARALLEL PROGRAMMING ENVIRONMENT AND ITS EFFICIENT IMPLEMENTATION APPROVED BY DISSERTATION COMMITTEE: Copyright by Peter William Newton 1993 A GRAPHICAL RETARGETABLE PARALLEL PROGRAMMING ENVIRONMENT AND ITS EFFICIENT IMPLEMENTATION by PETER WILLIAM NEWTON, B.S., M.S.C.S |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/03-LOT.ps.Z, *19931115* xiii List of Tables Table 1-1. CODE Goals and Key Technologies. 5 Table 3-1. The Graph s Formal Parameters. 37 Table 9-1. CODE and HeNCE Compared. 168 |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/04-Intro.ps.Z, *19931115* 1 Chapter 1. Introduction Parallel computer architectures capable of supporting coarse-grain macro-level parallel computations have many attractive properties. They can be built economically from commodity components, and they have the most potential to combine breadth of applicability and scalability |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/14-Conc.ps.Z, *19931115* 173 Chapter 11. Conclusion The CODE 2.0 graphical parallel programming environment addresses the problem of effectively programming MIMD-style parallel computers at subroutine level granularities. To be effective for such machines, a programming language or environment must satisfy three practical |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/08-ExPg.ps.Z, *19931115* 57 Chapter 5. Example Programs and Performance Results In this chapter we will present several example parallel programs that were written using the CODE system. These programs form the basis for an experimental evaluation of CODE which concentrates on performance. They are also good examples of CODE |

| ftp://cs.utk.edu/pub/newton/code2/newton_diss/16-UCcode.ps.Z, *19931115* 183 Appendix B. Example Sequent Translation of UC Node This appendix contains a listing of the program text created for a sim- ple Unit of Computation node by the Sequent Symmetry translator. The node simple accepts an integer on an incoming arc and passes it out on an outgoing arc. See Figure B-1. |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f4.ps.Z, *19931116* 10002000300040005000t678910aAverage Fitness of Population |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f6.ps.Z, *19931116* 10002000300040005000t3456HEntropy of Denotation Matrix |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f5.ps.Z, *19931116* 10002000300040005000t45505560aAverage Fitness of Population |

| ftp://cs.utk.edu/pub/maclennan/SEECC.ps.Z, *19931116* Synthetic Ethology and Communication 1 DRAFT Not to be Reproduced or Generally Distributed Submitted to Adaptive Behavior Synthetic Ethology and the Evolution of Cooperative Communication Bruce J. MacLennan* & Gordon M. Burghardt *Computer Science Department University of Tennessee, Knoxville TN 37996, |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f1.ps.Z, *19931116* c / a c / a c / a c / a c / a c / a 1 c / a 3 1 42 11 12 12 21 1 1 |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f2.ps.Z, *19931116* offspringparent 2parent 1 C b A c b a C B A q h 1 |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f7.ps.Z, *19931116* 10002000300040005000t464850aAverage Fitness of Population |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f3.ps.Z, *19931116* 10002000300040005000t6.26.46.66.8aAverage Fitness of Population |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-195.ps.Z, *19931117* THE DEVELOPMENT AND IMPLEMENTATION OF A PERFORMANCE DATABASE SERVER Brian Howard LaRose Computer Science Department CS-93-195 August 1993 The evelopment and Implementation of a erformance atabase Server A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Brian |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-191.ps.Z, *19931117* SOFTWARE DISTRIBUTION USING XNETLIB 12 LaRose, Sharon Lewis, Keith Moore, Andrew Pearson, Jon Richardson, Bill Rosener, and Andrea Van Hull all made valuable contributions to the design, development, and testing of various versions of xnetlib. We also thank the many users of early versions of xnetlib |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-196.ps.Z, *19931117* THE DESIGN, IMPLEMENTATION AND PERFORMANCE OF A QUEUE MANAGER FOR PVM Douglas J. Sept Computer Science Department CS-93-196 August 1993 The esign, Implementation and erformance of a ueue anager for A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Douglas J. Sept |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-213.ps.Z, *19931117* PUBLIC INTERNATIONAL BENCHMARKS FOR PARALLEL COMPUTERS PARKBENCH Committee: Report-1 assembled by Roger Hockney (chairman) and Michael Berry (secretary) Computer Science Department University of Tennessee CS-93-213 November 1993 Public International Benchmarks for Parallel Computers PARKBENCH Committee: |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-197.ps.Z, *19931117* PARALLEL MAP ANALYSIS ON THE CM-5 FOR LANDSCAPE ECOLOGY MODELS Karen Stoner Minser Computer Science Department CS-93-197 August 1993 Parallel ap nalysis on the -5 for Landscape cology odels A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Karen Stoner Minser |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-194.ps.Z, *19931117* SVDPACKC (Version 1.0) USER'S GUIDE Michael Berry, Theresa Do, Gavin O'Brien Vijay Krishna, and Sowmini Varadhan Computer Science Department CS-93-194 April 1993 (Revised October 1993) SVDPACKC (Version 1.0) User's Guide1 Michael Berry 2 Theresa Do2 Gavin O'Brien2 Vijay Krishna2 Sowmini Varadhan2 |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-207.ps.Z, *19931117* DATA-PARALLEL IMPLEMENTATIONS OF MAP ANALYSIS AND ANIMAL MOVEMENT FOR LANDSCAPE ECOLOGY MODELS Ethel Jane Comiskey Computer Science Department CS-93-207 August 1993 Data-Parallel Implementations of Map Analysis and Animal Movement for Landscape Ecology Models A Thesis Presented for the Master of Science |

| ftp://cs.utk.edu/pub/maclennan/SHPCC-abs.ps.Z, *19931209* Massive Parallelism in Neurocomputation Extended Abstract Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu December 8, 1993 1 Introduction Many applications require the flexible processing of large amounts of ambiguous, incomplete, or redundant |

| ftp://cs.utk.edu/pub/pozo/papers/sparse.ps.Z, *19931216* A Sparse Matrix Library in C++ for High Performance Architectures Jack Dongarraxz, Andrew Lumsdaine , Xinhiu Niu Roldan Pozoz, Karin Remingtonx xOak Ridge National Laboratory zUniversity of Tennessee University of Notre Dame Mathematical Sciences Section Department of Computer Science Dept. of Computer |

| ftp://cs.utk.edu/pub/cs594-unp/utlogin.ps, *19940107* How to use a computer science account by modem from your computer at home You will need a home computer, communication software1, and a modem. ffl Terminal Setting | To connect to UTK computer network, you need to have the correct terminal setting on your machine. Parity should be set to NONE. Stop bit |

| ftp://cs.utk.edu/pub/maclennan/IS-tr.ps.Z, *19940125* Image and Symbol Continuous Computation and the Emergence of the Discrete Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville Dec. 18, 1992; rev. Aug. 5, 1993 |

| ftp://cs.utk.edu/pub/maclennan/GAC.ps.Z, *19940125* Grounding Analog Computers Bruce J. MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996 Email: maclennan@cs.utk.edu 1.0 Is Cognition Discrete or Continuous 1.1 Although I hate to haggle over words, Harnad's use of `analog' confuses a number of issues. The problem begins |

| ftp://cs.utk.edu/pub/btouranc/SurveyMonitoring.ps, *19940125* Monitoring of Distributed Memory Multicomputer Programs Maurice van Riek Laboratoire de l'Informatique du Parall elisme, CNRS-URA 1398 Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France. Bernard Tourancheau Department of Computer Science, University of Tennessee 107 Ayres Hall, Knoxville, TN |

| ftp://cs.utk.edu/pub/maclennan/SEECC-f8.ps.Z, *19940126* 200040006000800010000t34567HEntropy of Denotation Matrix |

| ftp://cs.utk.edu/pub/pozo/papers/lapack++_guide.ps.Z, *19940202* LAPACK++ V. 0.9 Users' Guide Janunary 1994 February 2, 1994 Jack Dongarra Roldan Pozo David Walker Oak Ridge National Laboratory University of Tennessee, Knoxville Note: This document is a working draft. Its contents are incomplete and subject to change. Updates will be available from netlib@onrl.gov. |

| ftp://cs.utk.edu/pub/pozo/papers/oonski.ps.Z, *19940216* An Object Oriented Design for High Performance Linear Algebra on Distributed Memory Architectures Jack J. Dongarraxz, Roldan Pozoz, and David W. Walkerx xOak Ridge National Laboratory zUniversity of Tennessee Mathematical Sciences Section Department of Computer Science P. O. Box 2008, Bldg. 6012 107 |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-120.ps.Z, *19940218* onstructive o le it 1 Karl Abrahamson1 Michael R. Fellows2 Michael A. Langston3 Bernard M.E. Moret4 1 e e e e e, e e , , e e e e e, e , , . . e e e e e, e e e ee, e, e e e e e, e e e , e e, str ct Powerful and widely applicable, yet inherently non-constructive, tools have recently become available for |

| ftp://cs.utk.edu/pub/TechReports/1992/ut-cs-92-176.ps.Z, *19940218* Ra ee o in an, ichael A. Langston and Si harthan Ramachan ramurthi Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 USA October 9, 1992 A stract In this paper, we present a practical approach for finding optimal solutions for fixed-parameter versions of many VLSI layout |

| ftp://cs.utk.edu/pub/TechReports/1992/ut-cs-92-175.ps.Z, *19940218* To Appear in Journal of Parallel Algorithms and Applications I ersus I Co utation: x erience it on- u eric arallel l orit s Clay P. Breshears and Michael A. Langston Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 s c c - c c c . c - c . c c c - , c . c c - , c c c c - c |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-98.ps.Z, *19940218* n=(log n) for optimality, we observe that our algorithms are also efficient in the usual sense (their speedup is within a polylogarithmic factor of optimal) for any value of n, suggesting that they may have practical merit even for relatively small files. As long as main memory remains a critical |

| ftp://cs.utk.edu/pub/TechReports/1991/ut-cs-91-126.ps.Z, *19940218* November, 1992 anc . innersle and Michael . angston bstract Gate matrix layout is a well-known -complete problem that arises at the heart of a number of VLSI layout styles. Despite its apparent general intractability, it has recently been shown that it can be solved in ( 2 ) time whenever the number of |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-102.ps.Z, *19940218* Accepted for Publication in I ransactions on o uters iao un uan and Michael A. Langston |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-106.ps.Z, *19940218* Accepted for Publication in The omputer ournal Bing-Chao Huangy and Michael A. Langstonz |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-107.ps.Z, *19940218* May 1990 s s s Michael R. Fellows and Michael A. Langston A stract. We have previously established the existence of decision algorithms with low-degree polynomial running times for a number of difficult combinatorial problems, including many that can be stated in terms of VLSI layout, placement, |

| ftp://cs.utk.edu/pub/TechReports/1989/ut-cs-89-94.ps.Z, *19940218* to appear, SIAM Journal on Discrete Mathematics Michael R. Fellows y and Michael A. Langston z |

| ftp://cs.utk.edu/pub/TechReports/1989/ut-cs-89-95.ps.Z, *19940218* Accepted for Publication in SI ournal on iscrete at ematics Donald K. Friesen and Michael A. Langston A stract. Consider the classic bin packing problem, in which one seeks to pack a list of items into the minimum number of unit-capacity bins. We investigate the worst-case performance of a compound bin |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-206.ps.Z, *19940228* A Computationally Universal Field Computer That is Purely Linear David H. Wolperty Bruce J. MacLennanz September 14, 1993 |

| ftp://cs.utk.edu/pub/moore/solaris-t-shirt.ps, *19940228* Solaris (Ask me why.) |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-219.ps.Z, *19940228* Faster Checkpointing with N + 1 Parity James S. Plank Kai Liy University of Tennessee Technical Report CS-93-219 December, 1993 Submitted to FTCS-24 |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-199.ps.Z, *19940228* Image and Symbol Continuous Computation and the Emergence of the Discrete Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville August 5, 1993 |

| ftp://cs.utk.edu/pub/pozo/papers/lapack++.ps.Z, *19940307* LAPACK++: A Design Overview of Object-Oriented Extensions for High Performance Linear Algebra Jack J. Dongarra xz, Roldan Pozoz , and David W. Walkerx xOak Ridge National Laboratory zUniversity of Tennessee Mathematical Sciences Section Department of Computer Science |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-212.ps.Z, *19940331* ritical xa ination f e c e a eore Michael D. Vose Computer Science Dept., 107 Ayres Hall The University of Tennessee Knoxville, TN 37996-1301 vose@cs.utk.edu A stract The classical schema theorem is examined with the aim of discerning what it really has to say. The conclusion reached is probably |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-189.ps.Z, *19940331* omputin tea tate istri utions it out ancellation1 MICHAEL D. VOSE C.S. Dept., 107 Ayres Hall The University of Tennessee Knoxville, TN 37996 (615) 974-5067 vose@cs.utk.edu bstract: The steady state distribution of a Markov chain is represented such that no cancellation occurs. The resulting form is |

| ftp://cs.utk.edu/pub/manchek/thesis.ps.Z, *19940511* DESIGN AND IMPLEMENTATION OF PVM VERSION 3 A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Robert J. Manchek May, 1994 ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Jack Dongarra, for the guidance he provided, and a job that was a lot of fun. I thank |

| ftp://cs.utk.edu/pub/plank/papers/FTCS24.1994.ps.Z, *19940513* Faster Checkpointing with N + 1 Parity James S. Plank Department of Computer Science Univeristy of Tennessee Knoxville, TN 37996 plank@cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu Appearing in: 24th Annual International Symposium on |

| ftp://cs.utk.edu/pub/maclennan/MPNC.ps.Z, *19940525* Massive Parallelism in Neurocomputation Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu May 23, 1994 |

| ftp://cs.utk.edu/pub/plank/thesis/part_1.ps.Z, *19940613* EFFICIENT CHECKPOINTING ON MIMD ARCHITECTURES James Steven Plank A DISSERTATION PRESENTED TO THE FACULTY OF PRINCETON UNIVERSITY IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY RECOMMENDED FOR ACCEPTANCE BY THE DEPARTMENT OF COMPUTER SCEINCE June 1993 c Copyright by James Steven Plank, 1993. All |

| ftp://cs.utk.edu/pub/plank/thesis/part_a.ps.Z, *19940613* ffl Staggering the disk writes is clearly a good idea, as it drastically reduces the processors' serialization at the disks. When the program's behavior is not characterized by synchronization or large messages, this staggering can be effected by the Network Sweeping algorithm much more so than the |

| ftp://cs.utk.edu/pub/plank/thesis/part_8.ps.Z, *19940613* SNSCLNS0123SecondsSMALL-INDSNSCLNS05101520MED-INDSNSCLNS0102030LARGE-INDSNSCLNS01234SecondsSMALL-SYNCSNSCLNS0510152025MED-SYNCSNSCLNS010203040LARGE-SYNC Figure 4.7: Checkpoint Overhead of Basic Algorithms. there are no mCL messages sent in the staggering version of the NS algorithm. Thus, the degree of |

| ftp://cs.utk.edu/pub/plank/thesis/part_b.ps.Z, *19940613* William J. Dally. Express cubes: Improving the performance of kary n-cube interconnection networks. IEEE Transactions on Computers, 40(9):1016023, September 1991. William J. Dally. Fine-grain concurrent computing. Research Directions in Computer Science: An MIT Perspective, pages |

| ftp://cs.utk.edu/pub/plank/thesis/part_9.ps.Z, *19940613* Non-compressed Compressed Instance Checkpoint Recovery Checkpoint Recovery of Time Time Time Time a m s (SNS) (SNS) SMALL-IND 3.76 (sec) 4.72 (sec) 2.92 (sec) 4.50 (sec) SMALL-SYNC 4.44 4.78 3.32 4.62 MED-IND 21.30 22.08 5.52 6.39 MED-SYNC 25.42 23.67 6.26 5.44 LARGE-IND 39.76 43.09 8.01 9.21 LARGE-SYNC |

| ftp://cs.utk.edu/pub/plank/thesis/part_c.ps.Z, *19940613* Heap Size Checkpoint Overhead (sec) (Mbytes) Sequential Main Memory CLL Copy-on-Write 0.789 3.53 2.34 0.60 1.40 2.985 10.28 6.12 3.84 4.42 3.962 12.76 7.25 3.36 3.85 4.938 15.73 9.70 3.63 5.66 7.868 24.67 16.39 4.31 6.08 9.821 30.03 19.16 4.38 7.79 Table B.2: Checkpoint Overhead of Multiprocessor |

| ftp://cs.utk.edu/pub/plank/thesis/part_5.ps.Z, *19940613* Chapter 3 Optimizations This chapter outlines several optimizations that can be made to the basic algorithms presented in Chapter 2. These optimizations are meant to improve the algorithms' performance on the three speed metrics described in the introduction: checkpoint time, overhead, and latency. |

| ftp://cs.utk.edu/pub/plank/thesis/part_6.ps.Z, *19940613* Normal OperationNormal MessageP1P2. . .WaitWaitWaitStaggered CheckpointingMarker MessageNormal MessageP1P2WaitWaitWaitCheckpointCheckpointNon-Staggered CheckpointingMarker MessageNormal MessageP1P2WaitWaitCheckpointWait + Checkpoint Figure 3.7: Staggering Checkpoints with a Synchronizing Pair of |

| ftp://cs.utk.edu/pub/plank/thesis/part_4.ps.Z, *19940613* 2.6.2 Sending mNS Messages on Wormhole Routing In- terconnection Networks Above we state that a processor p sends a mNS marker along a link when it determines that no more cross-cut messages will be sent along that link. However, we do not state specifically how p makes this determination. The example |

| ftp://cs.utk.edu/pub/plank/thesis/part_3.ps.Z, *19940613* We omit the proof of this, as it may be found in . The crux of the proof is that any cross-cut message must arrive at a node after it has checkpointed, but before it receives mCL from the sender. Thus they are all identified and logged properly. One way to measure the complexity of this algorithm |

| ftp://cs.utk.edu/pub/plank/thesis/part_2.ps.Z, *19940613* For example, suppose processors P1 and P2 do not synchronize with each other when they checkpoint. Then the following scenario could occur: Time P1 P2 1 Save CPU State 2 Read x 3 Write x 4 Save CPU State 5 Read x The resulting checkpoint will be invalid regardless of when the state of memory is saved. |

| ftp://cs.utk.edu/pub/plank/thesis/part_d.ps.Z, *19940613* Algorithm Main Com- Stag- Checkpoint Checkpoint Memory pressed gered Time (sec) Overhead (sec) SNS No No - 45.52 45.61 SNS Yes No - 49.12 42.25 SNS No Yes - 9.87 9.90 SNS Yes Yes - 11.70 7.61 CL No No No 43.47 43.55 CL No No Yes 46.76 46.68 CL Yes No No 46.77 41.38 CL Yes No Yes 50.75 43.48 CL No Yes No |

| ftp://cs.utk.edu/pub/plank/thesis/part_7.ps.Z, *19940613* 326496128Page Size (KBytes)02004006008001000Number of Page FaultsNumber of Faults vs. Page Size0326496128Page Size (KBytes)0.00.20.40.60.81.01.21.4Maximum Fault Time (Sec)Maximum Fault Time vs. Page Size Figure 4.4: Results of Altering the Page Size. Thus, even at the beginning of the checkpoint, when |

| ftp://cs.utk.edu/pub/rowan/thesis/thesis.ps.Z, *19940621* FUNCTIONAL STABILITY ANALYSIS OF NUMERICAL ALGORITHMS APPROVED BY SUPERVISORY COMMITTEE: Copyright by Thomas Harvey Rowan 1990 FUNCTIONAL STABILITY ANALYSIS OF NUMERICAL ALGORITHMS by THOMAS HARVEY ROWAN, B.A., M.S. DISSERTATION Presented to the Faculty of the Graduate School of The University of Texas |

| ftp://cs.utk.edu/pub/rowan/papers/xnetlib.ps.Z, *19940621* - 15 - 161. Anthony Skjellum, Lawrence Livermore National Laboratory, 7000 East Ave., L- 316, P.O. Box 808 Livermore, CA 94551 162. Danny C. Sorensen, Department of Mathematical Sciences, Rice University, P.O. Box 1892, Houston, TX 77251 163. G. W. Stewart, Computer Science Department, University of |

| ftp://cs.utk.edu/pub/rowan/papers/netlib_services.ps.Z, *19940621* ORNL/TM-12680 Engineering Physics and Mathematics Division Mathematical Sciences Section NETLI SER ICES AND RESOURCES Shirley . Browne Jack J. Dongarra Stan C. Green Keith Moore Thomas H. Rowan Reed C. Wade Department of Computer Science University of Tennessee 107 Ayres Hall Knoxville, TN 37996-1301 |

| ftp://cs.utk.edu/pub/plank/thesis/thesis.ps.Z, *19940712* 178 BIBLIOGRAPHY Tankqueray, D., Snelling, D. F., A Distriubted Memory Implementation of the Shallow Water Equations," Dawn of Massively Parallel Processing in Meterology, eds. Hoffman, G.R., Maretis, D. K., Springer-Verlag, 1990. Thinking Machines Corporation, Communications Architecture in |

| ftp://cs.utk.edu/pub/bampton/Thesis/thesis.ps.Z, *19940728* Solving Imperfect Information Games Using the Monte Carlo Heuristic A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Howard James Bampton August 1994 Contents 1 Introduction 1 1.1 Why study games : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : |

| ftp://cs.utk.edu/pub/plank/papers/CS-94-243.ps.Z, *19940805* A Failure Correction Technique for Parallel Storage Devices with Minimal Device Overhead James S. Plank Department of Computer Science Univeristy of Tennessee Knoxville, TN 37996 plank@cs.utk.edu Joel Friedman Department of Mathematics Univeristy of British Columbia Vancouver, BC V6T 172 Canada |

| ftp://cs.utk.edu/pub/plank/papers/SHPCC94.ps.Z, *19940808* Performance Results of ickp | A Consistent Checkpointer on the iPSC/860 James S. Plank Department of Computer Science Univeristy of Tennessee Knoxville, TN 37996 plank@cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu Appearing in: Scalable |

| ftp://cs.utk.edu/pub/plank/papers/TPDS-94.ps.Z, *19940809* Low-Latency, Concurrent Checkpointing for Parallel Programs James S. Plank Department of Computer Science Univeristy of Tennessee Knoxville, TN 37996 plank@cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu appearing in: IEEE Transactions on |

| ftp://cs.utk.edu/pub/plank/papers/IJPP-92.ps.Z, *19940826* An Efficient Checkpointing Method for Multicomputers with Wormhole Routing Kai Li Jeffrey F. Naughtony James S. Plank In International Journal of Parallel Programming, Vol. 20, No. 3, June, 1992, pp 159-180 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-242.ps.Z, *19940909* Libckpt: Transparent Checkpointing under Unix James S. Plank Micah Beck Gerry Kingsley Department of Computer Science 107 Ayres Hall University of Tennessee Knoxville, TN 37996 @cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-185.ps.Z, *19940909* LAPACK WORKING NOTE 56 REDUCING COMMUNICATION COSTS IN THE CONJUGATE GRADIENT ALGORITHM ON DISTRIBUTED MEMORY MULTIPROCESSORS EDUARDO D'AZEVEDOy , VICTOR EIJKHOUTz AND CHARLES ROMINE y |

| ftp://cs.utk.edu/pub/TechReports/1992/ut-cs-92-169.ps.Z, *19940909* LAPACK working note 50 Distributed Sparse Data Structures for Linear Algebra Operations Victor Eijkhout Department of Computer Science University of Tennessee, Knoxville eijkhout@cs.utk.edu September 15, 1992 |

| ftp://cs.utk.edu/pub/TechReports/1992/ut-cs-92-170.ps.Z, *19940909* LAPACK working note 51 Qualitative Properties of the Conjugate Gradient and Lanczos Methods in a Matrix Framework Victor Eijkhout University of Tennessee Department of Computer Science 107 Ayres Hall, Knoxville, TN 37996-1301 eijkhout@cs.utk.edu September 15, 1992 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-243.ps.Z, *19940909* A Failure Correction Technique for Parallel Storage Devices with Minimal Device Overhead James S. Plank Department of Computer Science Univeristy of Tennessee Knoxville, TN 37996 plank@cs.utk.edu Joel Friedman Department of Mathematics Univeristy of British Columbia Vancouver, BC V6T 172 Canada |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-245.ps.Z, *19940912* Blocked Data Distribution for the Conjugate Gradient Algorithm on the CRAY T3D Michael W. Berryy Charles Grasslz Vijay K. Krishnax |

| ftp://cs.utk.edu/pub/TechReports/1993/ut-cs-93-jgraph-usenix.ps.Z, *19940916* Jgraph A Filter for Plotting Graphs in PostScript James S. Plank Princeton University |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-247.ps.Z, *19940920* Vector Processing on Scalar Architectures Micah Beck and Antonio Castellanos Department of Computer Science University of Tennessee, Knoxville TN 37996-1301 fbeck, castellag@cs.utk.edu September 1994 |

| ftp://cs.utk.edu/pub/plank/jgraph/alb.ps, *19941003* (a)(b) |

| ftp://cs.utk.edu/pub/plank/jgraph/col6.ps, *19941003* 20004000600080001000012345678Compilation Activity |

| ftp://cs.utk.edu/pub/plank/jgraph/acc.ps, *19941003* 20406080100Percentage of bus utilizationDBQSGERNG |

| ftp://cs.utk.edu/pub/plank/jgraph/fbf.ps, *19941003* UniversityofTennesseePrinceton102030405040302010UniversityFootballBudgets=1 GazillionDollars |

| ftp://cs.utk.edu/pub/plank/jgraph/wortman.ps, *19941003* 1000200030004000Time (clock ticks)12345678ProcessorLexical AnalysisImportSplitterParse/Dcl AnalysisStmt ProcessingMerge |

| ftp://cs.utk.edu/pub/plank/jgraph/mab2.ps, *19941003* Thu 00:00 Thu 06:00 Thu 12:00 Thu 18:00 Fri 00:00 Fri 06:00 Fri 12:00 Fri 18:00 Sat 00:00 Sat 06:00 Sat 12:00 Sat 18:00 Sun 00:00 Sun 06:00 Sun 12:00 Sun 18:00 Mon 00:00 Mon 06:00 Mon 12:00 Mon 18:00 Tue 00:00 Tue 06:00 Tue 12:00 Tue 18:00 Wed 00:00 Wed 06:00 Wed 12:00 Wed 18:00 |

| ftp://cs.utk.edu/pub/plank/jgraph/p15.ps, *19941003* Plank & Li: SHPCC '94: Performance results of ickp15No optimizationsMain memoryResults: The OptimizationsTimeOverheadCompressedMain mem., compressed0246Seconds0246Genetic:212K40K (89.9%)0246810Seconds0246810Sparse:317K77K (75.5%)0510152025Seconds0510152025FFT:1,015K907K (10.6%) |

| ftp://cs.utk.edu/pub/plank/jgraph/rej11.ps, *19941003* Plank & Li: SHPCC '94: Performance results of ickp113 Consistent Ckpt Algorithms- #3: NETWORK SWEEPING (NS): Good: No system freezing Bad: Allows a message state Good: O(n log n) markersNode 0Node 1Node 2Node 3 |

| ftp://cs.utk.edu/pub/plank/jgraph/usa.ps, *19941003* newgraphxaxis size 3 min 0 max 41 mhash 1 (* Put 1 tick between hash marks *) hash_labels font Times-Italic label : Qualifications...yaxis size 1.5 min 0 nodraw (* Don't draw the y-axis *)newcurve marktype ybar marksize 0 .6 fill .9 pts 41 4 35 3 17 2 14 1newstring hjl vjc (* These define justification |

| ftp://cs.utk.edu/pub/plank/jgraph/p25.ps, *19941003* Plank & Li: FTCS-24: Faster Checkpointing with N+1 Parity25Checkpoint Time and Overhead- Checkpoint Time & Overhead:- Compressing Diffs: Can be done very fast- However, size of buffers determines the frequency of checkpointing: If the buffer is too small, the application may spend too much time |

| ftp://cs.utk.edu/pub/plank/jgraph/incimp.ps, *19941003* MATSOLVECELLWATERMCNF-20020406080100Percentage reductionin checkpointsize-20020406080100Percentage reductionin checkpointsizeMATSOLVECELLWATERMCNF-20020406080100Percentage reductionin checkpointoverhead-20020406080100Percentage reductionin checkpointoverhead |

| ftp://cs.utk.edu/pub/plank/jgraph/col1.ps, *19941003* JimKaiJeffHistory |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-258.ps.Z, *19941010* Information Management Tools for Updating an SVD-Encoded Indexing Scheme Gavin W. O`Brien Computer Science Department CS-94-258 October 1994 Information Management Tools for Updating an SVD-Encoded Indexing Scheme A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-259.ps.Z, *19941010* Cross-Language Information Retrieval Using Latent Semantic Indexing Paul G. Young Computer Science Department CS-94-259 October 1994 Cross-Language Information Retrieval Using Latent Semantic Indexing A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Paul G. |

| ftp://cs.utk.edu/pub/newton/code2/wet94.ps.Z, *19941018* Chapter 1 Visual Programming and Parallel Computing Peter Newton |

| ftp://cs.utk.edu/pub/newton/misc/ut-cs-94-229.ps.Z, *19941018* Visual Programming and Parallel Computing 1 Visual Programming and Parallel Computing James C. Browne Jack Dongarra Syed I. Hyder Keith Moore Peter Newton |

| ftp://cs.utk.edu/pub/newton/code2/CodeICS92.ps.Z, *19941018* 1 The CODE 2.0 Graphical Parallel Programming Language* Peter Newton James C. Browne Department of Computer Sciences University of Texas at Austin Austin, TX 78712 |

| ftp://cs.utk.edu/pub/newton/code2/SC93tut.ps.Z, *19941018* Handouts Rounded Rectangle 1 1 SC 93 Graph/Vis Tutorial 1 Graphical/Visual Abstract Models and Tools in Parallel Computation J.C. Browne Peter Newton University of Texas at Austin Dept. of Computer Sciences Tutorial Supercomputing 93 2 SC 93 Graph/Vis Tutorial 2 Contents Brief Introduction to CODE 2.0 |

| ftp://cs.utk.edu/pub/maclennan/GRSTVI.ps.Z, *19941025* Gabor Representations of Spatiotemporal Visual Images Bruce MacLennan Technical Report CS-91-144 September 13, 1991; revised October 21, 1994 |

| ftp://cs.utk.edu/pub/newton/tmp/Exp_Code_Hence.ps.Z, *19941026* Experiences with CODE and HeNCE in Visual Programming for Parallel Computing 1 Experiences with CODE and HeNCE in Visual Programming for Parallel Computing James C. Browne Jack Dongarra Syed I. Hyder Keith Moore Peter Newton |

| ftp://cs.utk.edu/pub/plank/papers/CS-94-242.ps.Z, *19941027* Libckpt: Transparent Checkpointing under Unix James S. Plank Micah Beck Gerry Kingsley Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu Appearing |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-257.ps.Z, *19941031* Efficient Computation of the Singular Value Decomposition with Applications to Least Squares Problems Ming Gu James Demmely Inderjit Dhillonz October 1, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-254.ps.Z, *19941103* The Performance of Finding Eigenvalues and Eigenvectors of Dense Symmetric Matrices on Distributed Memory Computers J. Demmel K. Stanleyy |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-264.ps.Z, *19941109* XLSI { A Graphical User Interface for a Conceptual Retrieval System Susan Clower Allen Computer Science Department CS-94-264 December 1994 XLSI - A Graphical User Interface for a Conceptual Retrieval System A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Susan |

| ftp://cs.utk.edu/pub/newton/vpe/docs/SC94tut.ps.Z, *19941110* VPE Computer Science Dept. University of Tennessee 1 of 16 VPE: A Visual Environment for Heterogeneous Message-Passing Parallel Computing Peter Newton Jack Dongarra New project - Goals similar to HeNCE s. Simplify writing, building, running and debugging heterogeneous parallel programs. GUI - based |

| ftp://cs.utk.edu/pub/manchek/pvm/sc94-slides.ps.Z, *19941111* 1 Installation and Troubleshooting ffl Frequently asked questions ffl Useful facts about PVM ffl Debugging methodology 2 Set PVM ROOT ffl Always set PVM ROOT environment variable to directory where PVM is installed. ffl It must be set in order to build or run programs ffl In versions 3.3 and later, it |

| ftp://cs.utk.edu/pub/maclennan/FCATF.ps.Z, *19941124* Field Computation: A Theoretical Framework for Massively Parallel Analog Computation Parts I { IV Bruce MacLennan Department of Computer Science University of Tennessee CS-90-100 February 1990 Revised December 4, 1991 |

| ftp://cs.utk.edu/pub/maclennan/CCED-tr.ps.Z, *19941128* Continuous Computation and the Emergence of the Discrete Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville MacLennan@CS.UTK.edu March 24, 1994 Over many years I have searched for the point where myth and science join. It was clear to me for a long time that the origins of |

| ftp://cs.utk.edu/pub/maclennan/LpCirc.ps.Z, *19941128* Lp-Circular Functions Bruce MacLennan Computer Science Department University of Tennessee, Knoxville May 1, 1992 |

| ftp://cs.utk.edu/pub/maclennan/CFD.ps.Z, *19941128* The Calculus of Functional Differences and Integrals Bruce MacLennan Department of Computer Science The University of Tennessee Knoxville, TN 37996-1301 CS-89-80 April 1989 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-266.ps.Z, *19941205* The Design of the Land-Use Change Analysis System (LUCAS): Part I { Graphical User Interface R.L. MacIntyre, B.C. Hazen & M.W. Berry Computer Science Department CS-94-266 December 1994 |

| ftp://cs.utk.edu/pub/beck/uvector/ut-cs-94-247.ps.Z, *19941205* Vector Processing on Scalar Architectures Micah Beck and Antonio Castellanos Department of Computer Science University of Tennessee, Knoxville TN 37996-1301 fbeck, castellag@cs.utk.edu September 1994 |

| ftp://cs.utk.edu/pub/plank/papers/CS-94-268.ps.Z, *19941212* Algorithm-Based Diskless Checkpointing for Fault Tolerant Matrix Operations James S. Plank y Youngbae Kim y Jack J. Dongarra yz y Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu z Mathematical Science Section Oak Ridge National Laboratory |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-256.ps.Z, *19941214* Thomason 1 December 13, 1994 Parallel PET Reconstruction by EM Iteration with Reduced Processor Communications Soeren P. Olesen ms, Jens Gregor phd, Michael G. Thomason phd Department of Computer Science University of Tennessee Knoxville, TN 37996 Gary T. Smith md Department of Radiology University of |

| ftp://cs.utk.edu/pub/maclennan/digh/thesis.ps.Z, *19941216* DESIGN AND IMPLEMENTATION OF PVM VERSION 3 A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Robert J. Manchek May, 1994 ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Jack Dongarra, for the guidance he provided, and a job that was a lot of fun. I thank |

| ftp://cs.utk.edu/pub/maclennan/CCTMPS-tr.ps.Z, *19941219* Continuous Computation: Taking Massive Parallelism Seriously1 B. J. MacLennan Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 mclennan@utkcs2.cs.utk.edu 1 Discrete vs. Continuous Computation Modern computation is rooted in the discrete. This is most obvious in the digital |

| ftp://cs.utk.edu/pub/rowan/thesis/ch5.ps.Z, *19941230* Chapter 5 The Subplex Method for Unconstrained Optimization The subplex1 method (SUBPLEX), described in this chapter, is a new method for unconstrained optimization. SUBPLEX is a generalization of the Nelder-Mead simplex method (NMS), a popular method for optimizing noisy functions. The description of |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-271.ps.Z, *19950106* A Case Study of Latent Semantic Indexing M.W. Berry, S.T. Dumais & A.T. Shippy Computer Science Department CS-95-271 January 1995 A Case Study of Latent Semantic Indexing1 Michael W. Berry Department of Computer Science 107 Ayres Hall University of Tennessee Knoxville, TN 37996-1301 berry@cs.utk.edu |

| ftp://cs.utk.edu/pub/shuford/terminal/dec_term_function_checklist.ps, *19950106* FN Sanitized Host Interface Functions Checklist 12-Jan-94 Introduction This section provides tables showing ANSI host interface control functions that should be supported by VT510 software terminal emulation. To simplify understanding the huge number of functions and corresponding rich interface, this |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-240.ps.Z, *19950112* Basic Concepts for Distributed Sparse Linear Algebra Operations Victor Eijkhout and Roldan Pozo University of Tennessee Department of Computer Science 107 Ayres Hall, Knoxville, TN 37996-1301 eijkhout@cs.utk.edu, pozo@cs.utk.edu August 4, 1994 1 Introduction We introduce basic concepts for describing |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-252.ps.Z, *19950112* Relative Perturbation Theory: (I) Eigenvalue Variations Ren-Cang Li July 25, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-232.ps.Z, *19950112* To the Graduate Council: I am submitting herewith a thesis written by Robert J. Manchek entitled Design and Implementation of PVM Version 3." I have examined the final copy of this thesis for form and content and recommend that it be accepted in partial fulfillment of the requirements for the degree of |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-260.ps.Z, *19950112* Solving Secular Equations Stably and Efficiently Ren-Cang Li Department of Mathematics University of California at Berkeley Berkeley, California 94720 April, 1993 Dedicated to B. N. Parlett and W. Kahan on the occasion of their 60th birthdays |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-244.ps.Z, *19950112* PARALLELIZING THE QR ALGORITHM FOR THE UNSYMMETRIC ALGEBRAIC EIGENVALUE PROBLEM: MYTHS AND REALITY GREG HENRY AND ROBERT VAN DE GEIJNy |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-269.ps.Z, *19950112* Compiler-Assisted Checkpointing1 Micah Beck, James S. Plank and Gerry Kingsley Department of Computer Science University of Tennessee, Knoxville, TN 37996 fbeck, plank, kingsleyg@cs.utk.edu |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-225.ps.Z, *19950112* A Serial Implementation of Cuppen's Divide and Conquer Algorithm for the Symmetric Eigenvalue Problem J. Rutter Report No. UCB/CSD 94/799 February 1994 Computer Science Division (EECS) University of California Berkeley, California 94720 A Serial Implementation of Cuppen's Divide and Conquer Algorithm |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-250.ps.Z, *19950112* LAPACK Working Note 82 Call Conversion Interface (CCI) for LAPACK/ESSL y Jack Dongarra and Michael Kolatis Department of Computer Science University of Tennessee Knoxville, Tennessee 37996-1301 October 3, 1994 1 Overview This document reviews the initial version of the Call Conversion Interface (CCI) |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-229.ps.Z, *19950112* Visual Programming and Parallel Computing 1 Visual Programming and Parallel Computing James C. Browne Jack Dongarra Syed I. Hyder Keith Moore Peter Newton |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-237.ps.Z, *19950112* LAPACK{Style Algorithms and Software for Solving the Generalized Sylvester Equation and Estimating the Separation between Regular Matrix Pairs Bo K agstr om and Peter Poromaa December, 1993 Revised April, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-253.ps.Z, *19950112* Relative Perturbation Theory: (II) Eigenspace Variations Ren-Cang Li July 25, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-255.ps.Z, *19950112* Computing Eigenspaces with Specified Eigenvalues of a Regular Matrix Pair (A; B) and Condition Estimation: Theory, Algorithms and Software Bo K agstr om and Peter Poromaa April, 1994 Revised September, 1994 UMINF-94.04,ISSN-0348-0542 (Also published as LAPACK Working Note xx) |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-221.ps.Z, *19950112* A Highly Parallel Algorithm for the Reduction of a Nonsymmetric Matrix to Block Upper-Hessenberg Form Michael W. Berryy Jack J. Dongarra y Youngbae Kimy February 6, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-249.ps.Z, *19950112* LAPACK Working Note 81 Quick Installation Guide for LAPACK on Unix Systems1 Jack Dongarra and Susan Ostrouchov Department of Computer Science University of Tennessee Knoxville, Tennessee 37996-1301 REVISED: VERSION 2.0, September 30, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-241.ps.Z, *19950112* Computational variants of the CGS and BiCGstab methods Victor Eijkhout University of Tennessee Department of Computer Science 107 Ayres Hall, Knoxville, TN 37996-1301 eijkhout@cs.utk.edu August 4, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-263.ps.Z, *19950112* Prototype of the National High-Performance Software Exchange Shirley Browne, Jack Dongarra, Stan Green, Keith Moore, Tom Rowan, and Reed Wade Computer Science Department University of Tennessee Geoffrey Fox and Ken Hawick Northeast Parallel Architectures Center Syracuse University January 3, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-239.ps.Z, *19950112* Algorithmic Bombardment for the Iterative Solution of Linear Systems: A Poly-Iterative Approach Richard Barretty, Michael Berryz, Jack Dongarrazx, Victor Eijkhoutz, and Charles Rominex. July 27, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-251.ps.Z, *19950112* Relative Perturbation Bounds for the Unitary Polar Factor Ren-Cang Li July 25, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-228.ps.Z, *19950112* On the Correctness of Parallel Bisection in Floating Point James W. Demmel Computer Science Division and Department of Mathematics University of California Berkeley, California 94720 Inderjit Dhillony Computer Science Division University of California Berkeley, California 94720 Huan Renz Department of |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-268.ps.Z, *19950112* Algorithm-Based Diskless Checkpointing for Fault Tolerant Matrix Operations James S. Planky Youngbae Kimy Jack J. Dongarrayz yDepartment of Computer Science, University of Tennessee zMathematical Science Section, Oak Ridge National Laboratory |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-231.ps.Z, *19950112* LAPACK Working Note 71 IBM RS/6000-550 & -590 Performance for Selected Routines in ESSL /LAPACK /NAGy/IMSLzx Jack Dongarra and Michael Kolatis Department of Computer Science University of Tennessee Knoxville, Tennessee 37996-1301 May 3, 1994 1 Overview This document contains performance results for |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-246.ps.Z, *19950112* ORNL/TM-12470 Engineering Physics and Mathematics Division Mathematical Sciences Section THE DESIGN AND IMPLEMENTATION OF THE SCALAPACK LU, QR, AND CHOLESKY FACTORIZATION ROUTINES Jaeyoung Choi x Jack J. Dongarra xy Susan Ostrouchov x Antoine P. Petitet x David W. Walker y R. Clint Whaley x x Department |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-230.ps.Z, *19950112* MPI: A Message-Passing Interface Standard Message Passing Interface Forum April 21, 1994 This work was supported in part by ARPA and NSF under grant ASC-9310330, the National Science Foundation Science and Technology Center Cooperative Agreement No. CCR-8809615, and by the Commission of the European |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-222.ps.Z, *19950112* Netlib Services and Resources Shirley Browne, Jack Dongarra, Eric Grosse, Stan Green, Keith Moore, Tom Rowan, and Reed Wade |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-226.ps.Z, *19950112* A Parallel Performance Study of Jacobi-like Eigenvalue Solution Makan Pourzandi Laboratoire de l'Informatique du Parall elisme, Unit e de Recherche Associ ee 1398 du CNRS Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France Tel. (+33) 72 72 85 03 Fax (+33) 72 72 80 80 e-mail: |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-234.ps.Z, *19950112* LAPACK Working Note 73 Basic Linear Algebra Communication Subprograms: Analysis and Implementation Across Multiple Parallel Architectures. R. Clint Whaley y June 10, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-233.ps.Z, *19950112* The Computation of Elementary Unitary Matrices R.B. Lehoucq April 20, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-236.ps.Z, *19950112* A Sparse Matrix Library in C++ for High Performance Architectures Jack Dongarraxz, Andrew Lumsdaine , Xinhiu Niu Roldan Pozoz, Karin Remingtonx xOak Ridge National Laboratory zUniversity of Tennessee University of Notre Dame Mathematical Sciences Section Dept. of Computer Science Dept. of Computer |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-261.ps.Z, *19950112* Overview of VPE: A Visual Environment for Message-Passing Parallel Programming 16 6.0 Building and Running Programs VPE programs can be built and run directly from the VPE environment. When the user picks one of the build options from the Build menu, VPE will generate C (or Fortran and C) code to |

| ftp://cs.utk.edu/pub/plank/papers/CS-94-269.ps.Z, *19950127* Compiler-Assisted Checkpointing Micah Beck James S. Plank Gerry Kingsley Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu December 12, 1994 University of Tennessee Technical Report CS-94-269 Submitted to 25th Annual International Symposium on |

| ftp://cs.utk.edu/pub/maclennan/DD.ps.Z, *19950201* The discomforts of dualism A Review of Roger Penrose's The Emperor's New Mind: Concerning Computers, Minds, and the Laws of Physics (Appears in Behavioral and Brain Sciences 13, 4, December 1990, 673{674) Bruce MacLennan Department of Computer Science, University of Tennessee, Knoxville, TN 37996-1301 |

| ftp://cs.utk.edu/pub/maclennan/WLIOW.ps.Z, *19950201* Words Lie in our Way" (Position Paper for Symposium, What is Computing ") Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu June 20, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-276.ps.Z, *19950216* A Test Suite for PVM Martin Do (mdo@cs.utk.edu) Jack Dongarra (dongarra@cs.utk.edu) Emmanuel Jeannot (jeannot@lactaire.ens-lyon.fr) Philip J. Mucci (mucci@cs.utk.edu) December, 1994 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-273.ps.Z, *19950216* The Spectral Decomposition of Nonsymmetric Matrices on Distributed Memory Parallel Computers Z. Bai , J. Demmely, J. Dongarraz, A. Petitetx, H. Robinson{, K. Stanleyk January 9, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-275.ps.Z, *19950216* ORNL/TM-12472 Engineering Physics and Mathematics Division Mathematical Sciences Section THE DESIGN OF A PARALLEL DENSE LINEAR ALGEBRA SOFTWARE LIBRARY: REDUCTION TO HESSENBERG, TRIDIAGONAL. AND BIDIAGONAL FORM Jaeyoung Choi x Jack J. Dongarra xy David W. Walker y x Department of Computer Science |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-277.ps.Z, *19950216* Possibilities for Active Messaging in PVM Philip J. Mucci mucci@cs.utk.edu and Jack Dongarra dongarra@cs.utk.edu December 1994 |

| ftp://cs.utk.edu/pub/rowan/papers/nhse_prototype.ps.Z, *19950217* Prototype of the National High-Performance Software Exchange Shirley Browne, Jack Dongarra, Stan Green, Keith Moore, Tom Rowan, and Reed Wade Computer Science Department University of Tennessee Geoffrey Fox and Ken Hawick Northeast Parallel Architectures Center Syracuse University January 3, 1995 |

| ftp://cs.utk.edu/pub/rowan/papers/nhse.ps.Z, *19950217* National HPCC Software Exchange Shirley Browne, Jack Dongarra, Stan Green, Keith Moore, Tom Rowan, and Reed Wade University of Tennessee Geoffrey Fox and Ken Hawick Syracuse University Ken Kennedy Rice University Jim Pool California Institute of Technology Rick Stevens Argonne National Laboratory |

| ftp://cs.utk.edu/pub/maclennan/FCS-Users-Guide.ps.Z, *19950220* Field Computer Simulator User's Guide Tomislav Goles Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 goles@cs.utk.edu December 1990, revised February 16, 1995 |

| ftp://cs.utk.edu/pub/maclennan/IPDN.ps.Z, *19950220* Information Processing in the Dendritic Net Bruce MacLennany CS-92-180 |

| ftp://cs.utk.edu/pub/genetic/Fractals/fractal.ps.Z, *19950224* Jenny Juliany & Michael D. Vose Computer Science Dept., 107 Ayres Hall The University of Tennessee Knoxville, TN 37996-1301 vose cs.utk.edu (615) 974-5067 s The original schema theorem (an inequality) has been replaced by an equality which determines the expected next generation for a simple genetic |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-282.ps.Z, *19950315* An Incremental Algorithm for Satisfying Hierarchies of Multi-way, Dataflow Constraints Brad Vander Zanden Computer Science Department University of Tennessee Knoxville, TN 37996 bvz@cs.utk.edu |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-272.ps.Z, *19950315* National HPCC Software Exchange Shirley Browne, Jack Dongarra, Stan Green, Keith Moore, Tom Rowan, and Reed Wade University of Tennessee Geoffrey Fox and Ken Hawick Syracuse University Ken Kennedy Rice University Jim Pool California Institute of Technology Rick Stevens Argonne National Laboratory |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-274.ps.Z, *19950315* An Introduction to the MPI Standard Jack J. Dongarra University of Tennessee and Oak Ridge National Laboratory Steve W. Otto Oregon Graduate Institute of Science & Technology Marc Snir IBM, T.J. Watson Research Center David Walker Oak Ridge National Laboratory January 17, 1995 1 Contents 1 Introduction |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-280.ps.Z, *19950315* LAPACK Working Note 93 Installation Guide for ScaLAPACK1 J. Choi, J. J. Dongarra, L. S. Ostrouchov, A. Petitet, and R. C. Whaley Department of Computer Science University of Tennessee Knoxville, Tennessee 37996-1301 and J. Demmel, I. Dhillon, and K. Stanley Computer Science Division University of |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-279.ps.Z, *19950315* Digital Software and Data Repositories for Support of Scientific Computing Ronald Boisvert National Institute of Standards and Technology Shirley Browney University of Tennessee Jack Dongarra University of Tennessee and Oak Ridge National Lab Eric Grosse AT&T Bell Laboratories |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-278.ps.Z, *19950315* Location-Independent Naming for Virtual Distributed Software Repositories Shirley Browney, Jack Dongarra, Stan Green, Keith Moore Theresa Pepin, Tom Rowan, and Reed Wade University of Tennessee Eric Grosse AT&T Bell Laboratories |

| ftp://cs.utk.edu/pub/bvz/papers/lapidary.ps.Z, *19950320* Demonstrational and Constraint-Based Techniques for Pictorially Specifying Application Objects and Behaviors Brad Vander Zanden Brad A. Myers Computer Science Department School of Computer Science University of Tennessee Carnegie Mellon University Knoxville, TN 37996 Pittsburgh, PA 15213 bvz@cs.utk.edu |

| ftp://cs.utk.edu/pub/bvz/papers/uist94.ps.Z, *19950320* Optimizing Toolkit-Generated Graphical Interfaces Bradley T. Vander Zanden Computer Science Department University of Tennessee Knoxville, TN 37996 Tel: 1-615-974-8175 E-mail: bvz@cs.utk.edu |

| ftp://cs.utk.edu/pub/bvz/papers/tochi94.ps.Z, *19950320* Integrating Pointer Variables into One-Way Constraint Models Brad Vander Zanden Brad A. Myers Pedro Szekely Dario A. Giuse Computer Science Department School of Computer Science USC/Information Sciences Institute University of Tennessee Carnegie Mellon University 4676 Admiralty Way Knoxville, TN 37996 |

| ftp://cs.utk.edu/pub/dongarra/crpc-tutorial.ps, *19950410* 1 Tutorial on Software Tools and Templates for Developing a Software Library Jack Dongarra University of Tennessee and Oak Ridge National Laboratory 2 Why Parallel Computers ffl Desire to solve bigger, more realistic applications problems. ffl Fundamental limits are being approached. ffl More cost |

| ftp://cs.utk.edu/pub/maclennan/ICTPN-tr.ps.Z, *19950411* The Investigation of Consciousness Through Phenomenology and Neuroscience Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-287.ps.Z, *19950421* Management of the NHSE { a Virtual Distributed Digital Library Shirley Browney University of Tennessee Jack Dongarra Tom Rowan University of Tennessee and Oak Ridge National Lab Ken Kennedy Rice University March 27, 1995 |

| ftp://cs.utk.edu/pub/newton/vpe/docs/HWC95_Talk.ps.Z, *19950421* VPE Computer Science Dept. University of Tennessee 1 of 10 VPE Integrated Visual Programming Environment for Heterogeneous Message-Passing Parallel Computing HCW 95 Peter Newton Jack Dongarra Univ. of Tennessee Computer Science Dept. Project Goal: Simplify heterogeneous message-passing parallel |

| ftp://cs.utk.edu/pub/plank/papers/Booth.ps, *19950508* A. K. Tsakalidis, AVL-trees for localized search," Lecture Notes in Computer Science 172, J. Paredaens, ed., Springer-Verlag, Berlin, 1984, pp. 473-485. J. Valdes, R. E. Tarjan, and E. Lawler, The recognition of series-parallel digraphs," SIAM J. Comput., 11 (1982), pp.298-313. 88 D. |

| ftp://cs.utk.edu/pub/newton/misc/Exp_Code_Hence.ps.Z, *19950529* Experiences with CODE and HeNCE in Visual Programming for Parallel Computing 1 Experiences with CODE and HeNCE in Visual Programming for Parallel Computing James C. Browne Jack Dongarra Syed I. Hyder Keith Moore Peter Newton |

| ftp://cs.utk.edu/pub/newton/vpe/docs/ornl.ps.Z, *19950529* VPE Computer Science Dept. University of Tennessee 1 of 27 VPE Integrated Visual Programming Environment for Heterogeneous Message-Passing Parallel Computing Peter Newton Jack Dongarra Univ. of Tennessee Computer Science Dept. VPE Computer Science Dept. University of Tennessee 2 of 27 Project Goals |

| ftp://cs.utk.edu/pub/newton/vpe/docs/ut-cs-94-261.ps.Z, *19950529* Overview of VPE: A Visual Environment for Message-Passing Parallel Programming 16 6.0 Building and Running Programs VPE programs can be built and run directly from the VPE environment. When the user picks one of the build options from the Build menu, VPE will generate C (or Fortran and C) code to |

| ftp://cs.utk.edu/pub/newton/vpe/docs/hcw95.ps.Z, *19950601* Overview of VPE: A Visual Environment for Message-Passing Peter Newton Jack Dongarra Computer Science Department University of Tennessee Knoxville, TN, 37996-1301 |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-238.ps.Z, *19950614* The Land-Use Change Analysis System (LUCAS) for Evaluating Landscape Management Decisions Michael W. Berryy Richard O. Flammz Brett C. Hazeny Rhonda L. MacIntyrey |

| ftp://cs.utk.edu/pub/rowan/papers/paranoia_handout.ps.Z, *19950615* 1 C&TD User Services November 1991 Description and Results of PARANOIA Tom Rowan Mathematics Sciences Section Your favorite machine has just aborted your computation, accusing you of telling it to compute the square root of a negative quantity. Rechecking your code convinces you that you are innocent: |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug95-slides2.ps.Z, *19950619* 1 Implementation of PVM 3.3 2 PVM System Overview ffl PVM daemon (pvmd) { One manages each host of virtual machine { Mainly a message router, also has kernel-like functions { Has message entry points where tasks request service { Inter-host point of contact { Authentication { Creates processes { |

| ftp://cs.utk.edu/pub/maclennan/CFS-UMLC-tr.ps.Z, *19950703* Continuous Formal Systems: A Unifying Model in Language and Cognition Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville July 3, 1995 1 Introduction The idea of a calculus or discrete formal system is central to traditional models of language, knowledge, logic, cognition |

| ftp://cs.utk.edu/pub/gradappl/grad1.ps, *19950703* THE UNIVERSITY OF TENNESSEE, KNOXVILLE GRADUATE APPLICATION FOR ADMISSION RETURN TO: Graduate Admissions and Records, 218 Student Services Building, UTK, Knoxville, Tennessee, 37996-0220 I understand that withholding information required in this application or giving false information may make me |

| ftp://cs.utk.edu/pub/gradappl/csgta1.ps, *19950703* COMPUTER SCIENCE DEPARTMENT, UNIVERSITY OF TENNESSEE, KNOXVILLE APPLICATION FORM FOR GRADUATE ASSISTANTSHIP Return to: Computer Science Dept, 107 Ayres Hall, UTK, Knoxville, Tennessee, 37996-1301 Applying for Academic Year:__________ Which semester _____________ Program: ____MS ___PhD Applying for which |

| ftp://cs.utk.edu/pub/maclennan/CFS-UMLC-2col.ps.Z, *19950705* Continuous Formal Systems: A Unifying Model in Language and Cognition Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville 1 Introduction The idea of a calculus or discrete formal system is central to traditional models of language, knowledge, logic, cognition and |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-293.ps.Z, *19950711* PARALLEL ORDERING USING EDGE CONTRACTION PADMA RAGHAVAN fTECHNICAL REPORT: CS-95-293g y |

| ftp://cs.utk.edu/pub/plank/papers/FTCS25.1995.ps.Z, *19950803* Algorithm-Based Diskless Checkpointing for Fault Tolerant Matrix Operations James S. Plank y Youngbae Kim y Jack J. Dongarra yz y Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu z Mathematical Science Section Oak Ridge National Laboratory |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-294.ps.Z, *19950811* Chebyshev tau - QZ Algorithm Methods for Calculating Spectra of Hydrodynamic Stability Problems* J.J. Dongarra(1), B. Straughan(2) ** & D.W. Walker(2) (1)Department of Computer Science, University of Tennessee, Knoxville, Tennessee 37996- 1301, U.S.A., and Mathematical Sciences Section, Computer Science |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-286.ps.Z, *19950811* SUMMA: Scalable Universal Matrix Multiplication Algorithm Robert A. van de Geijn Department of Computer Sciences The University of Texas at Austin Austin, Texas 78712 rvdg@cs.utexas.edu Jerrell Watts Scalable Concurrent Programming Laboratory California Institute of Technology Pasadena, California 91125 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-291.ps.Z, *19950811* Reverse Communication Interface for Linear Algebra Templates for Iterative Methods Jack Dongarra University of Tennessee and Oak Ridge National Laboratory Victor Eijkhout University of California, Los Angeles Ajay Kalhan University of Tennessee May 18, 1995 1 Introduction In this report we describe a |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-289.ps.Z, *19950811* Modeling the Benefits of Mixed Data and Task Parallelism Soumen Chakrabarti James Demmely Katherine Yelick |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-281.ps.Z, *19950811* LAPACK Working Note 94 A User's Guide to the BLACS v1.0 Jack J. Dongarra, y R. Clint Whaley z June 7, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-290.ps.Z, *19950811* LAPACK++ V. 1.0 High Performance Linear Algebra Users' Guide April 1994 Jack Dongarra Roldan Pozo David Walker Oak Ridge National Laboratory University of Tennessee, Knoxville Contents 1 Introduction 3 2 Overview 5 2.1 Contents of LAPACK++ : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-292.ps.Z, *19950811* LAPACK Working Note 100 A Proposal for a Set of Parallel Basic Linear Algebra Subprograms J. Choiy, J. Dongarraz, S. Ostrouchovx, A. Petitetx, D. Walker{, and R. C. Whaleyx May, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-297.ps.Z, *19950811* TOP500 Supercomputer Sites Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Science Section Oak Ridge National Laboratory Oak Ridge, TN 37831-6367 dongarra@cs.utk.edu Hans W. Meuer Computing Center University of Mannheim D-68131 Mannheim |

| ftp://cs.utk.edu/pub/plank/jgraph/j.ps, *19950815* UniversityofTennesseePrinceton102030405040302010UniversityFootballBudgets=1 GazillionDollars |

| ftp://cs.utk.edu/pub/jepeway/NAF-95-atl/ip.ps.Z, *19950815* IP Operations: A Survival Guide Ron Tipton, Network Engineer Network Services University Of Tennessee and Chris Jepeway, Lab Engineer Computer Science Dept University of Tennessee, Knoxville September 27, 1995 Philosophy Our combined IP experience of 15 years has shown us that certain technical details |

| ftp://cs.utk.edu/pub/jepeway/NAF-95-atl/iptr.ps.Z, *19950815* IP Troubleshooting: Red Adair and You Ron Tipton, Network Engineer Network Services University Of Tennessee and Chris Jepeway, Lab Engineer Computer Science Dept University of Tennessee, Knoxville September 27, 1995 Format of The IP Troubleshooting Presentation I. Introduction II. Philosophy III. IP IV. |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-298.ps.Z, *19950821* The Design of the Land-Use Change Analysis System (LUCAS): Part II { Parallel and Distributed Implemetation B.C. Hazen & M.W. Berry Computer Science Department CS-95-298 August 1995 The Design of the Land-Use Change Analysis System (LUCAS): Part II - Parallel and Distributed Implementation B.C. Hazen |

| ftp://cs.utk.edu/pub/plank/scd/CONTOUR.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the CONTOUR program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) |

| ftp://cs.utk.edu/pub/plank/scd/WATER.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the WATER program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) |

| ftp://cs.utk.edu/pub/plank/scd/MCNF.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the MCNF program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) (MB) |

| ftp://cs.utk.edu/pub/plank/scd/LOG.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the LOG program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) (MB) |

| ftp://cs.utk.edu/pub/plank/scd/CELL.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the CELL program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) (MB) |

| ftp://cs.utk.edu/pub/plank/scd/SOLVE.ps.Z, *19950822* Raw data for the paper: Compressed Differences: An Algorithm for Fast Incremental Checkpointing by James S. Plank, Jian Xu and Robert H. B. Netzer Data for the SOLVE program Summary of Results General Info Algorithm Sbuffer # # Running Std. Checkpoint (MB) of of Time Dev. Size Tests Ckps (sec) (sec) |

| ftp://cs.utk.edu/pub/plank/papers/CS-95-302.ps.Z, *19950822* Compressed Differences: An Algorithm for Fast Incremental Checkpointing James S. Plank y Jian Xu z Robert H. B. Netzer z y Department of Computer Science University of Tennessee Knoxville, TN 37996 plank@cs.utk.edu z Computer Science Department Box 1910 Brown University Providence, RI 02912 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-300.ps.Z, *19950829* Parallelization of the Hoshen-Kopelman Algorithm Using a Finite State Machine M.W. Berry, J.M. Constantin & B.T. Vander Zanden Computer Science Department CS-95-300 August 1995 Submitted to Int. J. of Supercomputer Applications and High Performance Computing, (1995) Parallelization of the |

| ftp://cs.utk.edu/pub/manchek/pvm/epvm95-slides2.ps.Z, *19950908* Why would I want to reg Customizing PVM to suit your environment Jim Pruyne Computer Sciences Department University of Wisconsin { Madison pruyne@cs.wisc.edu http://www.cs.wisc.edu/ condor 1 Getting our Hooks into PVM ffl Condor - a distributed batch system Schedules user jobs onto machines (user |

| ftp://cs.utk.edu/pub/manchek/pvm/epvm95-slides3.ps.Z, *19950908* 1 Implementation of PVM 3.3 2 Evolution of PVM ffl Version 1 { Written during 1989-90 at Oak Ridge National Laboratory { Used internally in several projects ffl Version 2 { Written during February 1991 at University of Tennessee Knoxville { Released March 1991 { Simpler { More robust ffl Version 3 { |

| ftp://cs.utk.edu/pub/manchek/pvm/epvm95-slides1.ps.Z, *19950908* 1 Advanced PVM Programming ffl Basic PVM programming (not covered here) { Message passing { Spawning processes { Compiling and running programs { Dealing with PVM ffl Advanced PVM programming { Improving performance { Adding fault tolerance { Building interactive applications { Customizing the PVM |

| ftp://cs.utk.edu/pub/langston/benchmarks.ps.Z, *19950919* Parallel Benchmarks and Comparison-Based Computing Clay P. Breshears Department of Computer Science University of Southern Mississippi Hattiesburg, MS 39406 USA Michael A. Langston Department of Computer Science University of Tennessee Knoxville, TN 37996 USA |

| ftp://cs.utk.edu/pub/langston/applications.ps.Z, *19950920* On Algorithmic Applications of the Immersion Order An Overview of Ongoing Work Presented at the Third Slovenian International Conference on Graph Theory Michael A. Langston and Barbara C. Plaut Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-305.ps.Z, *19950928* On Algorithmic Applications of the Immersion Order An Overview of Ongoing Work Presented at the Third Slovenian International Conference on Graph Theory Michael A. Langston and Barbara C. Plaut Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-306.ps.Z, *19950928* Continuous Formal Systems: A Unifying Model in Language and Cognition Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville July 3, 1995 1 Introduction The idea of a calculus or discrete formal system is central to traditional models of language, knowledge, logic, cognition |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-307.ps.Z, *19950928* The Investigation of Consciousness Through Phenomenology and Neuroscience Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu |

| ftp://cs.utk.edu/pub/youngbae/PVM3/thesis.ps.gz, *19950929* DESIGN AND IMPLEMENTATION OF PVM VERSION 3 A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Robert J. Manchek May, 1994 ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Jack Dongarra, for the guidance he provided, and a job that was a lot of fun. I thank |

| ftp://cs.utk.edu/pub/youngbae/PVM3/epvm95-slides1.ps.gz, *19950929* 1 Advanced PVM Programming ffl Basic PVM programming (not covered here) { Message passing { Spawning processes { Compiling and running programs { Dealing with PVM ffl Advanced PVM programming { Improving performance { Adding fault tolerance { Building interactive applications { Customizing the PVM |

| ftp://cs.utk.edu/pub/youngbae/PVM3/epvm95-slides3.ps.gz, *19950929* 1 Implementation of PVM 3.3 2 Evolution of PVM ffl Version 1 { Written during 1989-90 at Oak Ridge National Laboratory { Used internally in several projects ffl Version 2 { Written during February 1991 at University of Tennessee Knoxville { Released March 1991 { Simpler { More robust ffl Version 3 { |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-267.ps.Z, *19951002* Parallel Benchmarks and Comparison-Based Computing Clay P. Breshears Department of Computer Science University of Southern Mississippi Hattiesburg, MS 39406 USA Michael A. Langston Department of Computer Science University of Tennessee Knoxville, TN 37996 USA |

| ftp://cs.utk.edu/pub/jics/I2PP.ps.Z, *19951009* Joint Institute for Computational Science Joint Institute for Computational Science, University of Tennessee, Knoxville, TN 37996, http://www-jics.cs.utk.edu / jics@cs.utk.edu Introduction to Parallel Processing Asim YarKhan yarkhan@cs.utk.edu Joint Institute for Computational Science jics@cs.utk.edu |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-296.ps.Z, *19951012* Computational Models of White-Tailed Deer in the Florida Everglades C.A. Abbott, M.W. Berry, E.J. Comiskey, L.J. Gross, & H.-K. Luh Computer Science Department CS-95-296 August 1995 Submitted to IEEE Computational Science & Engineering, (1995) Computational Models of White-Tailed Deer in the Florida |

| ftp://cs.utk.edu/pub/cowell/tlpd/tlpd.ps.gz, *19951023* Contents 9 Lives to Wonder 1 Madame Guillotine (track 1) : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 1 On Another Shore (track 2) : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : : 1 Softly Softly (track 3) : : : : : : : |

| ftp://cs.utk.edu/pub/ghenry/incoming/soe.ps, *19951030* A Parallel Algorithm for the Sylvester Observer Equation Christian H. Bischof Mathematics and Computer Science Division Argonne National Laboratory Argonne, IL 60439 Biswa Nath Dattay Department of Mathematical Sciences Northern Illinois University Dekalb, IL 60115 Avijit Purkayasthaz Department of |

| ftp://cs.utk.edu/pub/ghenry/hqr/hqr20.ps, *19951101* PARALLELIZING THE QR ALGORITHM FOR THE UNSYMMETRIC ALGEBRAIC EIGENVALUE PROBLEM: MYTHS AND REALITY GREG HENRY AND ROBERT VAN DE GEIJNy |

| ftp://cs.utk.edu/pub/plank/papers/USENIX-95W.ps.Z, *19951101* Libckpt: Transparent Checkpointing under Unix James S. Plank Micah Beck Gerry Kingsley Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu Kai Li Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu Appearing |

| ftp://cs.utk.edu/pub/maclennan/CFS-UMLC.ps, *19951102* Continuous Formal Systems: A Unifying Model in Language and Cognition Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville July 3, 1995 1 Introduction The idea of a calculus or discrete formal system is central to traditional models of language, knowledge, logic, cognition |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-304.ps.Z, *19951106* A Supernodal Approach to Sparse Partial Pivoting James W. Demmel Stanley C. Eisenstaty John R. Gilbertz Xiaoye S. Li Joseph W. H. Liux July 10, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-299.ps.Z, *19951106* Message-Passing Performance of Various Computers y Jack J. Dongarraz Tom Dunigan x |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-283.ps.Z, *19951106* LAPACK Working Note 95 ScaLAPACK: A Portable Linear Algebra Library for Distributed Memory Computers - Design Issues and Performance J. Choiy, J. Demmelz, I. Dhillonz, J. Dongarrax, S. Ostrouchovy, A. Petitety, K. Stanleyz, D. Walker{, and R. C. Whaleyy |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-301.ps.Z, *19951106* The Performance of PVM on MPP Systems Henri Casanova Jack Dongarra y Weicheng Jiang August 9, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-295.ps.Z, *19951106* LAPACK Working Note 101 A Proposal for a Fortran 90 Interface for LAPACK Jack J. Dongarra Jeremy Du Crozy Sven Hammarlingy Jerzy Wa sniewskiz Adam Zem lax August 10, 1995 1 Introduction The purpose of this paper is to initiate discussion of the design of a Fortran 90 interface to LAPACK . Our |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-303.ps.Z, *19951106* IML++ v. 1.2 Iterative Methods Library Reference Guide May 1995 Jack Dongarra Andrew Lumsdaine Roldan Pozo Karin Remington National Institute of Standards and Technology Oak Ridge National Laboratory University of Notre Dame University of Tennessee, Knoxville Contents 1 Introduction 2 2 Requirements 4 3 |

| ftp://cs.utk.edu/pub/bvz/papers/CS-95-282.ps.Z, *19951110* An Incremental Algorithm for Satisfying Hierarchies of Multi-way, Dataflow Constraints Brad Vander Zanden Computer Science Department University of Tennessee Knoxville, TN 37996 bvz@cs.utk.edu |

| ftp://cs.utk.edu/pub/kingsley/papers/suif-paper.ps.Z, *19951115* Compiler-Assisted Checkpoint Optimization Using SUIF Gerry Kingsley Micah Beck James S. Plank Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu] November 15, 1995 Submitted to First SUIF Compiler Workshop Compiler-Assisted Checkpoint Optimization |

| ftp://cs.utk.edu/pub/moore/thesis-951120.ps, *19951121* Chapter 1 Introduction 1.1 Introduction to the Problem Formanypeople in business, industry,and academia, electronic mail has become an essential communications tool. No longer a curiosity for its own sake, ``e-mail''isanestablished service in every networked of ce environment. Users depend on it, much |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-270.ps.Z, *19951211* Using Linear Algebra for Intelligent Information Retrieval M.W. Berry, S.T. Dumais & G.W. O'Brien Computer Science Department CS-94-270 December 1994 USING LINEAR ALGEBRA FOR INTELLIGENT INFORMATION RETRIEVAL MICHAEL W. BERRYy , SUSAN T. DUMAISz AND GAVIN W. O'BRIENx |

| ftp://cs.utk.edu/pub/TechReports/1994/ut-cs-94-262.ps.Z, *19951218* Parallel Models of Animal Migration in Northern Yellowstone National Park Ember Uziel & Michael W. Berry Computer Science Department CS-94-262 November 1994 Submitted to The InternationalJournal of SupercomputingApplicationsand High PerformanceComputing, (1995) Parallel Models of Animal Migration in |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-284.ps.Z, *19960125* Numerical Linear Algebra with Applications, Vol. 1(1), 1{27 (1996) Low-Rank Orthogonal Decompositions for Information Retrieval Applications M.W. Berry and R.D. Fierro Computer Science Department CS-95-284 April 1995 1070{5325/96/010001{27$18.50 Received 25 April 1995 c 1996 by John Wiley & Sons, Ltd. |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-319.ps.Z, *19960129* Estimating the Largest Singular Values/Vectors of Large Sparse Matrices via Modified Moments Sowmini Varadhan Computer Science Department CS-96-319 February 1996 Estimating the Largest Singular Values/Vectors of Large Sparse Matrices via Modified Moments A Dissertation Presented for the Doctor of |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-310.ps.Z, *19960131* Array Redistribution in ScaLAPACK using PVM Jack Dongarra The University of Tennessee and Oak Ridge National Laboratory Loic Prylli, Cyril Randriamaro and Bernard Tourancheauy LIP, Ecole Normale Suprieure de Lyon, France |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-309.ps.Z, *19960131* ISSN 1360-1725 UMIST Stability of the Diagonal Pivoting Method with Partial Pivoting N. J. Higham Numerical Analysis Report No. 265 July 1995 Manchester Centre for Computational Mathematics Numerical Analysis Reports DEPARTMENTS OF MATHEMATICS Reports available from: Department of Mathematics University |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-311.ps.Z, *19960131* 1 Templates for Linear Algebra Problems Zhaojun Bai University of Kentucky David Day University of Kentucky James Demmel University of California - Berkeley Jack Dongarra University of Tennessee and Oak Ridge National Laboratory Ming Gu University of California - Berkeley Axel Ruhe Chalmers University |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-316.ps.Z, *19960131* GEMM{Based Level 3 BLAS: Installation, Tuning and Use of the Model Implemen- tations and the Performance Evaluation Benchmark Bo K agstr om Per Ling Charles Van Loan y October 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-312.ps.Z, *19960131* Assessment of the NHSE Software Submission and Review Process Shirley Browne Tom Rowan October 23, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-315.ps.Z, *19960131* GEMM{Based Level 3 BLAS: High-Performance Model Implementations and Performance Evaluation Benchmark Bo K agstr om Per Ling Charles Van Loan y October 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-313.ps.Z, *19960131* NetSolve: A Network Server for Solving Computational Science Problems Henri Casanova Jack Dongarra y November 27, 1995 |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-308.ps.Z, *19960131* ISSN 1360-1725 UMIST Iterative Refinement and LAPACK N. J. Higham Numerical Analysis Report No. 277 September 1995 Manchester Centre for Computational Mathematics Numerical Analysis Reports DEPARTMENTS OF MATHEMATICS Reports available from: Department of Mathematics University of Manchester Manchester |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-317.ps.Z, *19960131* LAPACK Working Note 109 BLAS Technical Workshop1 Jack Dongarra2, Sven Hammarling3 and Susan Ostrouchov4 December 12, 1995 1This workshop was hosted by the University of Tennessee with partial support from the NSF S&T Center for Research on Parallel Computation. 2The University of Tennessee, Knoxville |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-318.ps.Z, *19960131* High Performance Computing in the U.S. in 1995 { An Analysis on the Basis of the TOP500 List Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Science Section Oak Ridge National Laboratory Oak Ridge, TN 37831-6367 dongarra@cs.utk.edu and Horst |

| ftp://cs.utk.edu/pub/maclennan/crumpton/crumpton.ps, *19960131* Evolution of Two Symbol Signals by Simulated Organisms A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Joseph J. Crumpton December 1994 |

| ftp://cs.utk.edu/pub/levy/SNARL/doc/snarl.ps, *19960202* SNARL v1.0 Reference Manual Simon D. Levy Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 levy@cs.utk.edu January 1996 1 Summary This document describes the data types and routines available in SNARL, a Simple Neural ARchitecture Library for C programmers. This document, |

| ftp://cs.utk.edu/pub/plank/papers/IEEETCOS95.ps.Z, *19960205* Compiler-Assisted Memory Exclusion for Fast Checkpointing James S. Plank Micah Beck Gerry Kingsley Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu Appearing in: IEEE Technical Committee on Operating Systems and Application Environments Special |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-320.ps.Z, *19960207* SNARL v1.0 Reference Manual Simon D. Levy Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 levy@cs.utk.edu January 1996 1 Summary This document describes the data types and routines available in SNARL, a Simple Neural ARchitecture Library for C programmers. This document, |

| ftp://cs.utk.edu/pub/genetic/foga4/paz.ps, *19960208* Towards a Theory of Deme Sizing for Multiple-Run Genetic Algorithms David E. Goldberg, Erick Cant u-Paz, Hillol Kargupta, and Jeffrey Horn Illinois Genetic Algorithms Laboratory University of Illinois at Urbana-Champaign 117 Transportation Building 104 S. Mathews Avenue Urbana, IL 61801 Office: (217) |

| ftp://cs.utk.edu/pub/genetic/foga4/culberson-2.ps, *19960208* On Searching ff-ary Hypercubes and Related Graphs: Extended Abstract Joseph Culberson Jonathan Lichtner January 30, 1996 1 Introduction In recent work there has been an attempt to understand genetic algorithms in terms of the searchscapes (or landscapes) they induce on various functions. In this |

| ftp://cs.utk.edu/pub/genetic/foga4/horn.ps, *19960208* Genetic Algorithms (with Sharing) in Search, Optimization, and Machine Learning1 Jeffrey Horn Department of Computer Science and The Illinois Genetic Algorithms Laboratory (IlliGAL) University of Illinois at Urbana-Champaign 117 Transportation Building, 104 South Mathews Avenue Urbana, Illinois |

| ftp://cs.utk.edu/pub/genetic/foga4/suzuki.ps, *19960208* Further Results on the Markov Chain Model of GAs and Their Application to SA-like Strategy Joe Suzuki (Information Systems Laboratory, Stanford University1) |

| ftp://cs.utk.edu/pub/genetic/foga4/mohan.ps, *19960208* Crossover via Probabilistic Inference (EXTENDED ABSTRACT) Chilukuri K. Mohan Kishan G. Mehrotra School of Computer and Information Science, 2-120 CST Syracuse University, NY 13244-4100 315-443-2368/(FAX)1122, ckmohan/mehrotra@syr.edu SUMMARY: This paper presents a new look at crossover operators, using |

| ftp://cs.utk.edu/pub/genetic/foga4/whigham.ps, *19960208* A Generalised Schema Theorem for Fixed and Variable-Length Structures P.A.Whigham Department of Computer Science, University College, University of New South Wales Australian Defence Force Academy Canberra ACT 2600 AUSTRALIA Email: paw@csadfa.cs.adfa.oz.au Phone: 06-268-8182 Fax: 06-268-8581 February 6, |

| ftp://cs.utk.edu/pub/genetic/foga4/dejong.ps, *19960208* Analyzing GAs Using Markov Models with Semantically Ordered States Kenneth A. De Jong and William M. Spears |

| ftp://cs.utk.edu/pub/genetic/foga4/mathias.ps, *19960208* Convergence Controlled Variation Larry J. Eshelman Keith Mathias J. David Schaffer Philips Research Laboratories Briarcliff Manor, NY 10510 lje/kem/ds1@philabs.philips.com 1 Introduction Forms of crossover that operate on pairs of binary strings have the important feature that allele values that are not |

| ftp://cs.utk.edu/pub/langston/k4algs.ps.Z, *19960208* February 1, 1996 Fast Algorithms for K4 Immersion Testing y Heather D. Booth, Rajeev Govindan, Michael A. Langston and Siddharthan Ramachandramurthi Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 |

| ftp://cs.utk.edu/pub/genetic/foga4/greenwood.ps, *19960208* Fitness Functions for Multipleobjective Optimization Problems: Combining Preferences With Pareto Rankings (EXTENDED ABSTRACT) Garrison W. Greenwood & Xiaobo (Sharon) Hu Dept. of Electrical & Computer Engineering Western Michigan University Kalamazoo, MI 49008 Joseph G. D'Ambrosio General Motors R&D |

| ftp://cs.utk.edu/pub/genetic/foga4/iba.ps, *19960208* Generating Random Trees for Genetic Programming Hitoshi IBA Machine Inference Section, Electrotechnical Laboratory (ETL), 1-1-4 Umezono, Tsukuba Science City, Ibaraki, 305, Japan. email. iba@etl.go.jp, |

| ftp://cs.utk.edu/pub/genetic/foga4/surry.ps, *19960208* A Formalism for Real-parameter Evolutionary Algorithms and Directed Recombination Patrick D. Surrya;b & Nicholas J. Radcliffea;b fpds,njrg@quadstone.co.uk aQuadstone Ltd, 16 Chester Street, Edinburgh, EH3 7RA, UK bDepartment of Mathematics, University of Edinburgh, The Kings Buildings, EH9 3JZ, UK |

| ftp://cs.utk.edu/pub/genetic/foga4/wright-2.ps, *19960208* A Search for Counterexamples to Two Conjectures on the Simple Genetic Algorithm Alden H. Wright and Garrett Bidwell Computer Science Dept. The University of Montana Missoula, MT 59812-1008 wright@cs.umt.edu and bidwell@cs.umt.edu |

| ftp://cs.utk.edu/pub/genetic/foga4/levenick.ps, *19960208* FOGA - 96 GA Directions J.R.Levenick From here, where Future directions for GA research. James R. Levenick Computer Science Willamette University Salem, OR. 97301 levenick@willamette.edu |

| ftp://cs.utk.edu/pub/genetic/foga4/goodman.ps, *19960208* Simultaneous Multi-Level Evolution 1 Simultaneous Multi-Level Evolution Gang Wang Erik D. Goodman Genetic Algorithm Research and Applications Group (GARAGe) Department of Computer Science and Case Center for Computer-Aided Engineering & Manufacturing Michigan State University Extended Abstract 1.0 |

| ftp://cs.utk.edu/pub/genetic/foga4/crawford.ps, *19960208* A Study of Fixed-Length Subset Recombination (Extended Abstract) Kelly D. Crawford Amoco Corporation kcrawford@amoco.com Roger L. Wainwright The University of Tulsa rogerw@penguin.mcs.utulsa.edu Cory J. Hoelting The University of Tulsa hoeltc@euler.mcs.utulsa.edu Dale A. Schoenefeld The University of |

| ftp://cs.utk.edu/pub/genetic/foga4/culberson-1.ps, *19960208* Schizophrenic Functions Joseph Culberson and Jonathan Lichtner January 29, 1996 1 Introduction In this paper we introduce the binary Schizophrenic function. This function has two classes of optima, and was constructed in such a way that searches using one-point mutation are expected to find one optima |

| ftp://cs.utk.edu/pub/genetic/foga4/rowe.ps, *19960208* The Generalised Royal Road Effect Jon Rowe Dept. of Computer and Information Sciences De Montfort University Hammerwood Gate Kents Hill Milton Keynes MK7 6HP United Kingdom jrowe@dmu.ac.uk |

| ftp://cs.utk.edu/pub/genetic/foga4/whitley.ps, *19960208* Nonlinearity, Walsh Coefficients, Hyperplane Ranking and the Simple Genetic Algorithm R. B. Heckendorn, D. Whitley, and S. Rana Department of Computer Science Colorado State University Fort Collins, Colorado 80523 USA fheckendo,whitley,ranag@cs.colostate.edu February 5, 1996 |

| ftp://cs.utk.edu/pub/genetic/foga4/geyers.ps, *19960208* Exact Uniform Initialization For Genetic Programming Walter B ohm, Andreas Geyer-Schulz Department of Mathematical Methods of Statistics, Institute of Statistics, Department of Applied Computer Science, Institute of Information Processing and Information Economics, Vienna University of Economics and |

| ftp://cs.utk.edu/pub/genetic/foga4/wright-1.ps, *19960208* Additively Decomposable Fitness Functions Richard K. Thompson Alden H. Wright rkt@charlo.math.umt.edu wright@cs.umt.edu Computer Science Dept. The University of Montana Missoula, MT 59812-1008 |

| ftp://cs.utk.edu/pub/genetic/foga4/mehrotra.ps, *19960208* Replicators, Majorization and Genetic Algorithms: New Models and Analytical tools (extended abstract) Anil Menon Kishan Mehrotra Chilukuri K. Mohan School of Computer & Information Science 2{120 Center for Science and Technology Syracuse University, Syracuse NY 13244-4100 |

| ftp://cs.utk.edu/pub/genetic/foga4/hart.ps, *19960208* A Stationary Point Convergence Theory for Evolutionary Algorithms (Extended Abstract) William E. Harty |

| ftp://cs.utk.edu/pub/genetic/foga4/harik.ps, *19960208* Learning Linkage Georges R. Harik Dept. of Electrical Engineering and Computer Science University of Michigan at Ann Arbor harikg@eecs.umich.edu David E. Goldberg Dept. of General Engineering University of Illinois at Urbana Champaign deg@uiuc.edu February 1, 1996 1 1 1 Introduction Early studies |

| ftp://cs.utk.edu/pub/genetic/foga4/shapiro.ps, *19960208* Genetic Algorithm Dynamics in Two-well Potentials with Basins and Barriers Jonathan Shapiro Department of Computer Science The University, Manchester M13 9PL U.K. Adam Pr ugel-Bennett NORDITA Blegdamsvej 17, DK-2100 Copenhagen O, Denmark 1 Introduction There are many ways in which optimization problems |

| ftp://cs.utk.edu/pub/genetic/foga4/rattray.ps, *19960208* Noisy Fitness Evaluation in Genetic Algorithms and the Dynamics of Learning L. Magnus Rattray Computer Science Department, University of Manchester, Oxford Road Manchester M13 9PL, U.K. January 22, 1996 1 Introduction The case where fitness evaluation is corrupted by noise is considered. This issue is |

| ftp://cs.utk.edu/pub/genetic/foga4/merkle.ps, *19960209* Optimal Parameter Selection for a Class of Evolutionary Algorithms (Extended Abstract) Laurence D. Merkle and Gary B. Lamont Department of Electrical and Computer Engineering Graduate School of Engineering Air Force Institute of Technology Wright-Patterson AFB, OH 45433 flmerkle, lamontg@afit.af.mil 1 |

| ftp://cs.utk.edu/pub/genetic/foga4/goldberg.ps, *19960211* What Should Genetic Algorithmists Do David E. Goldberg Department of General Engineering University of Illinois at Urbana-Champaign Urbana, IL 61801 1 An Unlikely Story Imagine you are waiting in Chicago O'Hare airport for your plane back home after a hectic workshop. The gate announcement is made and |

| ftp://cs.utk.edu/pub/genetic/foga4/graf.ps, *19960211* Phenotypes and Genotypes in Simulated Evolution Jeanine Graf ICD- Joseph von Fraunhofer 20 D-44227 Dortmund, Germany |

| ftp://cs.utk.edu/pub/genetic/foga4/vitanyi.ps, *19960211* A Discipline of Evolutionary Programming (Preliminary Draft Abstract) Paul Vit anyi CWI and University of Amsterdam February 7, 1996 |

| ftp://cs.utk.edu/pub/genetic/foga4/dasgupta.ps, *19960213* Representation and Operators in Genetic Search Cezary Z. Janikow & Dipankar Dasgupta Department of Mathematics and Computer Science University of Missouri | St. Louis St. Louis, MO 63121 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-321.ps.Z, *19960216* February 1, 1996 Fast Algorithms for K4 Immersion Testing y Heather D. Booth, Rajeev Govindan, Michael A. Langston and Siddharthan Ramachandramurthi Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 |

| ftp://cs.utk.edu/pub/genetic/foga4/aizawa.ps, *19960220* Fitness Landscape Characterization by Between Variance of Decompositions Akiko Aizawa National Center for Science Information Systems 3-29-1 Otsuka, Bunkyo-ku Tokyo 112, JAPAN akiko@rd.nacsis.ac.jp |

| ftp://cs.utk.edu/pub/genetic/foga4/reeves.ps, *19960220* Crossover and Recombination: Isolating the Building Blocks of a Genetic Algorithm Christian H ohn Institut f ur Grundlagen Elektrotechnik/Elektronik Technische Universit at Dresden, Germany Colin Reeves School of Mathematical and Information Sciences Coventry University UK Email: CRReeves@coventry.ac.uk |

| ftp://cs.utk.edu/pub/genetic/foga4/schwehm-parallel.ps, *19960220* Parallel Population Models for Genetic Algorithms Markus Schwehm Universit at Erlangen-N urnberg, IMMD VII Martensstr. 3, D-91058 Erlangen, Germany |

| ftp://cs.utk.edu/pub/genetic/foga4/schwehm-global.ps, *19960220* Global Optimization by Generational Methods Markus Schwehm University of Erlangen-N urnberg, IMMD VII Martensstr. 3, D-91058 Erlangen, Germany |

| ftp://cs.utk.edu/pub/genetic/foga4/bornholdt.ps, *19960220* Probing GA performance of fitness landscapes Stefan Bornholdt Institut f ur Theoretische Physik, Universit at Kiel Leibnizstr. 15, D-24098 Kiel, Germany bornholdt@theo-physik.uni-kiel.de January 31, 1996 The dynamics of populations in genetic algorithms can be approximated by a cumulant expansion of the |

| ftp://cs.utk.edu/pub/genetic/foga4/tkammeye.ps, *19960220* Stochastic Context-Free Grammar Induction with a Genetic Algorithm Using Local Search Thomas E. Kammeyer and Richard K. Belew {tkammeye,rik}@cs.ucsd.edu Cognitive Computer Science Research Group Computer Science and Engineering Dept. Mail Code 0114 University of California, San Diego 9500 Gilman Drive |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1f.ps.Z, *19960223* PVM User Group Meeting 1996 Beginner Tutorial 10 Systems That Work With PVM ffl Programming Tools and Libraries tkpvm { Graphics, Tcl + PVM ScaLAPACK { Parallel linear algebra routines Pade { Supports development of applications Pious { Parallel filesystem VPE { Visual programming environment (prototype |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1h.ps.Z, *19960223* PVM User Group Meeting 1996 Beginner Tutorial 11 How to Get PVM ffl PVM home page URL (Oak Ridge) is http://www/epm/ornl/gov/pvm/pvm home.html ffl PVM source code, user's guide, examples and related material are published on Netlib, a software repository with several sites around the world. { To get |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1b.ps.Z, *19960223* Thousands of Existing PVM ApplicationsThousands of applications have already been parallelized using PVMThese include Commercial applications such as Dyna3d and many more internal business applications. Survey results showed PVM applications in all the following areas and more... Meteorology Contaminant |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1c.ps.Z, *19960223* PVM Use by BusinessThe following is just a sample of the many companies using PVM today: Hundreds of companies are using PVM to solve important scientific, industrial, and medical problems. The following industries have saved millions of dollars by using this software as a cost-effective way to design |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1a.ps.Z, *19960223* 1 Beginner PVM Tutorial PVM User Group Meeting Santa Fe, NM 25 February, 1996 PVM User Group Meeting 1996 Beginner Tutorial 2 Topics ffl Introduction ffl Overview of PVM ffl History ffl Unique features in PVM ffl PVM user survey ffl XPVM ffl 3rd party systems for use with PVM ffl How to get and install |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1g.ps.Z, *19960223* Combining the Power of Distributed Computing with the Visualization Power of AVSPVM / AVSInteractive VisualizationComputational SteeringCollaborationVisualization DataSteeringPVM / AVSThe PVM/AVS interface easily couples a parallel computation to state-of-the-art visualizationWhat data is sent |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s1d.ps.Z, *19960223* PVM User Group Meeting 1996 Beginner Tutorial 7 Programming in PVM ffl A simple message-passing environment { Hosts, Tasks, Messages { No enforced topology { Virtual machine can be composed of any mix of machine types ffl Process Control { Tasks can be spawned/killed anywhere in the virtual machine ffl |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s2a.ps.Z, *19960223* 1 Advanced PVM Programming Tutorial PVM User Group Meeting Santa Fe, NM 25 February, 1996 PVM User Group Meeting 1996 Advanced Tutorial 2 Advanced Topics ffl Improving performance { Choosing pack and send options { Large amounts of data { Load balancing ffl Packing derived data types ffl Programming |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/s2b.ps.Z, *19960223* Why would I want to reg Customizing PVM to suit your environment Jim Pruyne Computer Sciences Department University of Wisconsin { Madison pruyne@cs.wisc.edu http://www.cs.wisc.edu/ condor 1 Getting our Hooks into PVM ffl Condor - a distributed batch system Schedules user jobs onto machines (user |

| ftp://cs.utk.edu/pub/jepeway/NAF-96-lv/ipts.ps, *19960301* February 27, 1996 IP Troubleshooting: Red Adair and You Ron Tipton, Network Engineer Network Services University Of Tennessee and Chris Jepeway, Lab Engineer Computer Science Dept University of Tennessee, Knoxville February 27, 1996 Format of The IP Troubleshooting Presentation I. Introduction II. |

| ftp://cs.utk.edu/pub/jepeway/NAF-96-lv/dns.ps, *19960301* February 29, 1996 Domain Name Service Administration: A Survival Guide Ron Tipton, Network Engineer Network Services University Of Tennessee and Chris Jepeway, Lab Engineer Computer Science Dept University of Tennessee, Knoxville February 29, 1996 What is DNS DNS software allows a client to ask of a |

| ftp://cs.utk.edu/pub/jepeway/NAF-96-lv/ip.ps, *19960301* February 25, 1996 IP Operations: A Survival Guide Ron Tipton, Network Engineer Network Services University Of Tennessee and Chris Jepeway, Lab Engineer Computer Science Dept University of Tennessee, Knoxville February 25, 1996 Philosophy Our combined IP experience of 15 years has shown us that certain |

| ftp://cs.utk.edu/pub/plank/papers/ADCKP.ps.Z, *19960301* Fault Tolerant Matrix Operations for Networks of Workstations Using Diskless Checkpointing James S. Plank y Youngbae Kim y Jack J. Dongarra yz y Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu z Mathematical Science Section Oak Ridge |

| ftp://cs.utk.edu/pub/manchek/pvm/pvmug96tut/examples.ps.Z, *19960304* 1 Five Examples ffl Inner product ffl Matrix vector multiply (row distribution) ffl Matrix vector multiply (column distribution) ffl Integration to evaluate ss ffl Solve 1-D heat equation 2 Inner Product Problem: In parallel compute s = nX i=1 xTy Solution: ffl Master - Slave ffl Uses the following PVM |

| ftp://cs.utk.edu/pub/plank/jgraph/c2j.ps, *19960307* #include main(){ printf("Hello World
");} |

| ftp://cs.utk.edu/pub/plank/jgraph/p07.ps.Z, *19960307* Plank, Kim & Dongarra: Algorithm-Based Diskless Checkpointing...: FTCS-25 (1995)07RunningReadyFailedLoaded |

| ftp://cs.utk.edu/pub/padma/CAPSS/Doc/capss.ps, *19960314* CAPSS: A CARTESIAN PARALLEL SPARSE SOLVER PADMA RAGHAVAN y |

| ftp://cs.utk.edu/pub/jics/PVM_guide.ps.Z, *19960323* A Beginner's Guide to PVM Parallel Virtual Machine JOINT INSTITUTE FOR COMPUTATIONAL SCIENCE Clay Breshears Joint Institute for Computational Science 104 South College University of Tennessee Knoxville, TN 37996{1301, USA Last Revision: March 23, 1996 Email Contact: jics@cs.utk.edu Author's Email: |

| ftp://cs.utk.edu/pub/wickman/audio-tape.ps, *19960329* From the Jamie Zawinski Collection of Fine Tapes Gang of 4EntertainmentSolid Gold Entertainment Solid Gold Paralysed What we All Want Why Theory If I could keep it for myself Outside the trains don't run on time Cheeseburger The Republic In the Ditch A Hole in the Wallet He'd send in the Army Ether |

| ftp://cs.utk.edu/pub/plank/origami/poly.ps, *19960404* Penultimate Polyhedra James S. Plank Department of Computer Science University of Tennessee 107 Ayres Hall Knoxville, TN 37996 plank@cs.utk.edu http://www.cs.utk.edu/~plank/plank/origami/origami.html March 28, 1996 Introduction These are some notes that I originally hacked up for my sister. They |

| ftp://cs.utk.edu/pub/TechReports/1995/ut-cs-95-296a.ps.Z, *19960416* Computational Models of White-Tailed Deer in the Florida Everglades C.A. Abbott, M.W. Berry, E.J. Comiskey, J.C. Dempsey, L.J. Gross, & H.-K. Luh Computer Science Department CS-95-296 August 1995 Submitted to IEEE Computational Science & Engineering, (1995) Computational Models of White-Tailed Deer in |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-324.ps.Z, *19960417* Key Concepts For Parallel Out-Of-Core LU Factorization Jack J. Dongarraz Sven Hammarling y David W. Walkerz March 28, 1996 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-325.ps.Z, *19960425* Overview of Recent Supercomputers Aad J. van der Steen and Jack J. Dongarra y |

| ftp://cs.utk.edu/pub/rowan/thesis/ch2.ps.Z, *19960428* Chapter 2 Theoretical Background This chapter covers the theoretical background necessary for the derivation of functional stability analysis. Key definitions are given and the concepts of conditioning and stability are discussed. The definitions given here will be used in the next chapter as the |

| ftp://cs.utk.edu/pub/rowan/thesis/bib.ps.Z, *19960503* BIBLIOGRAPHY M. J. Box. A comparison of several current optimization methods, and the use of transformations in constrained problems. Computer Journal, 9:677, 1966. P. A. Businger. Algorithm 253: Eigenvalues of a real symmetric matrix by the QR method. CACM, 8(4):21718, 1965. A. K. |

| ftp://cs.utk.edu/pub/rowan/thesis/ch7.ps.Z, *19960503* Chapter 7 Numerical Results The program INSTAB is a single-precision, FORTRAN 77 implementation of functional stability analysis. It is coded using the BLAS (Basic Linear Algebra Subprograms). INSTAB also uses SUBPLX, the program that implements the subplex optimization method. Although used by INSTAB, |

| ftp://cs.utk.edu/pub/rowan/thesis/ch3.ps.Z, *19960503* Chapter 3 Functional Stability Analysis The theoretical background presented in Chapter 2 will be used in this chapter for the derivation of functional stability analysis, a new method for detecting instability in numerical algorithms. This chapter first reviews previous approaches for analyzing the |

| ftp://cs.utk.edu/pub/rowan/thesis/ch8.ps.Z, *19960503* Chapter 8 Discussion 8.1 Summary of Results In Chapter 7, the program INSTAB, implementing functional stability analysis, tested the stability of a variety of numerical algorithms. To provide a rigorous test of INSTAB's capabilities, the tests for instability used a simple methodology that took no |

| ftp://cs.utk.edu/pub/rowan/thesis/ch4.ps.Z, *19960503* Chapter 4 Estimating a Lower Bound on the Backward Error The implementation of functional stability analysis consists of two parts. The first part derives a function B L that estimates a lower bound on the backward error. Each estimate is obtained while treating the numerical algorithm as a black box. |

| ftp://cs.utk.edu/pub/rowan/thesis/ch0.ps.Z, *19960503* FUNCTIONAL STABILITY ANALYSIS OF NUMERICAL ALGORITHMS APPROVED BY SUPERVISORY COMMITTEE: Copyright by Thomas Harvey Rowan 1990 FUNCTIONAL STABILITY ANALYSIS OF NUMERICAL ALGORITHMS by THOMAS HARVEY ROWAN, B.A., M.S. DISSERTATION Presented to the Faculty of the Graduate School of The University of Texas |

| ftp://cs.utk.edu/pub/rowan/thesis/ch6.ps.Z, *19960503* Chapter 6 Computational Details Chapter 4 derived a function B L that estimates a lower bound on the backward error. Chapter 5 described an optimization method for maximizing B L. This chapter discusses some of the computational details involved in combining the evaluation and maximization of B L to |

| ftp://cs.utk.edu/pub/rowan/thesis/ch1.ps.Z, *19960503* Chapter 1 Introduction 1.1 Detecting Instability In Numerical Algorithms If a stable numerical algorithm is applied to a problem whose true solution is insensitive to small input perturbations, the computed solution is guaranteed to be accurate. If an unstable numerical algorithm is applied to the same |

| ftp://cs.utk.edu/pub/gradappl/csgta2.ps, *19960507* GRADUATE ASSISTANTSHIP APPLICATION FORM, PAGE 2 Computer Science Department, UT Knoxville For applicants for research assistantships: You need to have been discussing this with a faculty member with grant/research funds. We do not have money set aside for general GRA positions. Which Professor |

| ftp://cs.utk.edu/pub/gradappl/grad2.ps, *19960507* UTK GRADUATE APPLICATION FORM, PAGE 2 List all colleges and universities (including UT) previously attended, dates and degrees received. One official transcript from all previous institutions (except UT Knoxville) must be sent directly to the Office of Graduate Admissions and Records. All transcripts |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab4/lab4.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lab 4 In this lab you will: ffl Review what you learned in Lab 3 about analyzing an algorithm by using experiments. ffl Learn more Common Lisp: the dotimes, do, and notany looping facilities; how to create and use arrays; and using functions |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab2/lab2.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lab 2 In this lab you will: ffl Review the Lisp functions you wrote for Lab 1, paying careful attention to style. ffl Begin investigation of the run-time of programs, and its relationship to inputs and their length. ffl Continue your practice |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab5/lab5.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lab 5 In this lab you will: ffl Review what you learned in Lab 4 about greedy algorithms. ffl Increase your familiarity with divide-and-conquer algorithms, by coding merge-sort. ffl Do an experiment to compare the run times of two sorting |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab6/lab6.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1994 Lab 6 In this lab you will: ffl Discover how to analyze merge-sort by developing and solving a recurrence equation. ffl Extend your experience with recurrence equations by analyzing several versions of quick-sort and a selection algorithm. |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab3/lab3.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lab 3 In this lab you will: ffl Review the Lisp functions you wrote for Lab 2, paying careful attention to style. ffl Discuss the analysis of an algorithm called select-sort, it being a loop within a loop, by: { Paper-and-pencil analysis. { |

| ftp://cs.utk.edu/pub/maclennan/cs320/lab1/lab1.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lab 1 In this lab you will: ffl Prepare your environment on the SUN workstations. ffl Practice some Lisp skills, including: { Evaluation. { Quoting. { List functions, including cons, car, cdr, member, and append. { Defining functions. { |

| ftp://cs.utk.edu/pub/maclennan/cs320/style.ps, *19960508* 1 Computer Science 320, Problem Solving Spring semester, 1995 Lisp style sheet A typical Lisp program is composed of many functions. By following certain recommendations when writing Lisp, you will create a program that is: ffl much more likely to be correct; ffl much easier for other readers to |

| ftp://cs.utk.edu/pub/maclennan/cs320/BPlab/lab3a.ps, *19960508* 1 CS 420 | Neural Networks Back-Propogation part 1 March 8, 1995 In this lab you will program the back-propogation algorithm for use in part 2 of this lab. As usual, you may program in either C or Lisp, and this part of the lab is due on Monday, March 27. Consider: programming back-propogation will be |

| ftp://cs.utk.edu/pub/maclennan/cs320/BPlab/lab3b.ps, *19960508* 1 CS 420 | Neural Networks Back-Propagation part 2 March 30, 1995 For this part of the lab you will be using the back-propagation network you designed in part one of this lab. This lab is due Monday, April 10th. On that date hand in this handout along with appropriate output and email your program to |

| ftp://cs.utk.edu/pub/TechReports/Thesis/Letsche.ps.Z, *19960509* Toward Large-Scale Information Retrieval Using Latent Semantic Indexing A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Todd A. Letsche August 1996 Acknowledgments I am deeply indebted to Dr. Michael Berry, my major advisor, for his kind guidance and support. I |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest15.ps, *19960528* BHESS STABILITY TESTS tbhess15rmstol1.grep tbhess15rmstol10.grep tbhess15rmstol100.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest10c.ps, *19960528* BHESS STABILITY TESTS tbhess10rmstol1c.grep tbhess10rmstol10c.grep tbhess10rmstol100c.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest5c.ps, *19960528* BHESS STABILITY TESTS tbhess5rmstol1c.grep tbhess5rmstol10c.grep tbhess5rmstol100c.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/jepeway/resume.ps, *19960528* Chris Markland Jepeway Home Work 3316 Timberlake107 Ayres Hall Knoxville, TN 37920 Knoxville, TN 37996 423-579-3376 423-974-5067 jepeway@cs.utk.edu University of Tennessee Knoxville, May 1994 Knoxville, TNEducation Masters of Science in Computer Science Rice University,1985 Houston, TX Bachelor of |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest5.ps, *19960528* BHESS STABILITY TESTS tbhess5rmstol1.grep tbhess5rmstol10.grep tbhess5rmstol100.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-328.ps.Z, *19960528* PVMPI: An Integration of the PVM and MPI Systems Graham E. Fagg Jack J. Dongarray April 12, 1996 |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest10.ps, *19960528* BHESS STABILITY TESTS tbhess10rmstol1.grep tbhess10rmstol10.grep tbhess10rmstol100.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/stabtest15c.ps, *19960528* BHESS STABILITY TESTS tbhess15rmstol1c.grep tbhess15rmstol10c.grep tbhess15rmstol100c.grep backward error estimate optimization trial # 1e-20 1e-19 1e-18 1e-17 1e-16 1e-15 1e-14 1e-13 1e-12 1e-11 1e-10 1e-09 1e-08 1e-07 1e-06 1e-05 1e-04 1e-03 1e-02 1e-01 1e+00 0.00 20.00 40.00 60.00 80.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-327.ps.Z, *19960530* A Focus-of-Attention EM-ML Algorithm for PET Reconstruction Jens Gregor and Dean A. Huff Department of Computer Science University of Tennessee 107 Ayres Hall Knoxville, TN 37996{1301 USA May 28, 1996 |

| ftp://cs.utk.edu/pub/sqrl/UME.ps, *19960626* Specification for the Usage Model Editor Software Quality Research Laboratory Computer Science University of Tennessee November 21, 1994 Contents 1 Introduction 5 2 User Interface Specification 6 2.1 Operating System/Application Interface : : : : : : : : : : : : 7 2.1.1 Description : : : : : : : : : : : |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-329.ps.Z, *19960718* Software Repository Interoperability Shirley Browne Jack Dongarra University of Tennessee Kay Hohn ASSET Tim Niesen Raytheon Technical Report UT-CS-96-329 July 1996 |

| ftp://cs.utk.edu/pub/plank/papers/CS-96-332.ps.Z, *19960719* A Tutorial on Reed-Solomon Coding for Fault-Tolerance in RAID-like Systems James S. Plank Technical Report UT-CS-96-332 Department of Computer Science University of Tennessee Knoxville, TN 37996 July 19, 1996 This a preliminary version submitted to Software, Practice & Experience." The paper is under |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-332.ps.Z, *19960719* A Tutorial on Reed-Solomon Coding for Fault-Tolerance in RAID-like Systems James S. Plank Technical Report UT-CS-96-332 Department of Computer Science University of Tennessee Knoxville, TN 37996 July 19, 1996 This a preliminary version submitted to Software, Practice & Experience." The paper is under |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/bhess_instab.ps, *19960727* More Words on INSTAB Testing of BHESS Tom July 26, 1996 |

| ftp://cs.utk.edu/pub/plank/papers/SRDS-15.ps.Z, *19960729* Improving the Performance of Coordinated Checkpointers on Networks of Workstations using RAID Techniques James S. Plank Department of Computer Science University of Tennessee Knoxville, TN 37996 plank@cs.utk.edu See http://www.cs.utk.edu/~plank/plank/papers/SRDS-15.html for information Appearing in: |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-333.ps.Z, *19960729* Block-Partitioned Algorithms for Solving the Linear Least Squares Problem Gregorio Quintana-Ort y Enrique S. Quintana-Ort z Antoine Petitet x LAPACK Working Note # 113. University of Tennessee at Knoxville Technical Report CS-96-333 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-334.ps.Z, *19960807* A BLAS-3 Version of the QR Factorization with Column Pivoting Gregorio Quintana-Ort y Xiaobai Sun z Christian H. Bischof x PRISM Working Note # 26 LAPACK Working Note # 114 Univ. of Tennessee at Knoxville Computer Sci. Dept. CS-96-334 August, 1996 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-326.ps.Z, *19960809* LAPACK Working Note 111, UTK, http://www.netlib.org/lapack/lawnsOptimizing Matrix Multiply using PHiPA C: a Portable,High-Performance, ANSI C Coding MethodologyJeff Bilmes , Krste Asanovi cy, Jim Demmelz, Dominic Lamx, Chee-Whye Chin{August 8, 1996AbstractBLAS3 operations hav e great potential for |

| ftp://cs.utk.edu/pub/plank/papers/CS-96-335.ps.Z, *19960809* Memory Exclusion: Optimizing the Performance of Checkpointing Systems James S. Plank y Yuqun Chen z Kai Li z Micah Beck y Gerry Kingsley y y Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu z Department of Computer Science Princeton University |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-335.ps.Z, *19960809* Memory Exclusion: Optimizing the Performance of Checkpointing Systems James S. Plank y Yuqun Chen z Kai Li z Micah Beck y Gerry Kingsley y y Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu z Department of Computer Science Princeton University |

| ftp://cs.utk.edu/pub/maclennan/FCMC.ps.Z, *19960812* FIELD COMPUTATION IN MOTOR CONTROL Bruce MacLennan Computer Science Department University of Tennessee, Knoxville (USA) |

| ftp://cs.utk.edu/pub/plank/libckpt/memex/fig3.ps, *19960812* I0ISTSWM0100200300Overhead perCheckpoint (sec)NormalCopy-on-write10%20%0IEIE0IEIENNET020406080NormalCopy-on-write5%0IEIE0IEIESOLVE050100NormalCopy-on-write5%0IEIE0IEIECELL020406080100Overhead perCheckpoint (sec)NormalCopy-on-write5%0I0IEIGEN020406080100NormalCopy-on-write5%0 -- No memory exclusionI -- |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-330.ps.Z, *19960828* LAPACK Working Note 112: Practical Experience in the Dangers of Heterogeneous Computing L. S. Blackford , A. Clearyy, J. Demmelz, I. Dhillonz, J. Dongarrax, S. Hammarling{, A. Petitety, H. Renz, K. Stanleyz, R. C. Whaleyy |

| ftp://cs.utk.edu/pub/maclennan/JCS2.ps, *19960830* THE ELEMENTS OF CONSCIOUSNESS AND THEIR NEURODYNAMICAL CORRELATES Bruce MacLennan,y Computer Science Department, University of Tennessee, Knoxville 37996, USA |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-331.ps.Z, *19960830* PROTOPHENOMENA AND THEIR NEURODYNAMICAL CORRELATES Bruce MacLennan, CS-96-331 July 18, 1996 |

| ftp://cs.utk.edu/pub/maclennan/PNC.ps, *19960830* PROTOPHENOMENA AND THEIR NEURODYNAMICAL CORRELATES Bruce MacLennan, CS-96-331 July 18, 1996 |

| ftp://cs.utk.edu/pub/plank/papers/Frontiers96.ps.Z, *19960901* Fault Tolerant Matrix Operations Using Checksum and Reverse Computation Youngbae Kimy James S. Planky Jack J. Dongarrayz yDepartment of Computer Science zMathematical Sciences Section University of Tennessee Oak Ridge National Laboratory Knoxville, TN 37996-1301 Oak Ridge, TN 37821-6367 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-337.ps.Z, *19960905* Field Computation in Motor Control Bruce MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu August 13, 1996 |

| ftp://cs.utk.edu/pub/rowan/projects/bhess/tm_template/tm.ps, *19960913* ORNL/TM-99999 Computer Science and Mathematics Division Mathematical Sciences Section TEST PAPER TO SHOW ORNL TECHREPORT FORMAT D. Scott McCrickard y Thomas H. Rowan z I. M. Another z y Graphics, Visualization, and Usability Center College of Computing Georgia Institute of Technology Atlanta, GA 30332 |

| ftp://cs.utk.edu/pub/bvz/papers/uist96.ps.Z, *19960930* An Empirical Study of Constraint Usage in Graphical Applications Bradley T. Vander Zanden Computer Science Department University of Tennessee Knoxville, TN 37996 bvz@cs.utk.edu Scott A. Venckus Computer Science Department University of Tennessee Knoxville, TN 37996 svenckus@cs.utk.edu |

| ftp://cs.utk.edu/pub/bvz/uist96.ps.Z, *19960930* An Empirical Study of Constraint Usage in Graphical Applications Bradley T. Vander Zanden Computer Science Department University of Tennessee Knoxville, TN 37996 bvz@cs.utk.edu Scott A. Venckus Computer Science Department University of Tennessee Knoxville, TN 37996 svenckus@cs.utk.edu |

| ftp://cs.utk.edu/pub/maclennan/SEASCf-tr.ps.Z, *19961008* Synthetic Ethology: An Approach to the Study of Communication CS-90-104 Bruce MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996-1301 internet address: maclennan@cs.utk.edu May 1990 |

| ftp://cs.utk.edu/pub/maclennan/ECPSMf.ps.Z, *19961008* Evolution of Communication in a Population of Simple Machines CS-90-99 Bruce MacLennan Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 maclennan@cs.utk.edu January 1990 |

| ftp://cs.utk.edu/pub/maclennan/SEASCf.ps.Z, *19961008* Synthetic Ethology: An Approach to the Study of Communication Bruce MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996-1301 internet address: maclennan@cs.utk.edu November 23, 1990 |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-99.ps.Z, *19961009* Evolution of Communication in a Population of Simple Machines CS-90-99 Bruce MacLennan Department of Computer Science University of Tennessee Knoxville, TN 37996-1301 maclennan@cs.utk.edu January 1990 |

| ftp://cs.utk.edu/pub/TechReports/1990/ut-cs-90-104.ps.Z, *19961009* Synthetic Ethology: An Approach to the Study of Communication CS-90-104 Bruce MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996-1301 internet address: maclennan@cs.utk.edu May 1990 |

| ftp://cs.utk.edu/pub/maclennan/MMD-tr.ps.Z, *19961022* -1- Mixing Memory and Desire: Want and Will in Neural Modeling* Bruce J. MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996-1301 Telephone: 423-974-5067 / Fax: 423-974-4404 E-Mail: maclennan@cs.utk.edu Homepage: http://www.cs.utk.edu/~mclennan |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-336.ps.Z, *19961029* On the Error Analysis and Implementation of Some Eigenvalue Decomposition and Singular Value Decomposition Algorithms by Huan Ren B.S. (Peking University, P.R.China) 1991 A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Applied Mathematics in |

| ftp://cs.utk.edu/pub/plank/classes/cs360/tests/t1/Tukey.ps, *19961029* 20406080100Score out of 100%020406080100Score out of 100%1234Total(10)(12)(14)(14)(50) |

| ftp://cs.utk.edu/pub/maclennan/FCMC-tr.ps, *19961030* Field Computation in Motor Control Bruce MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu Revised October 30, 1996 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-339.ps.Z, *19961105* -1- Mixing Memory and Desire: Want and Will in Neural Modeling* Bruce J. MacLennan Computer Science Department University of Tennessee Knoxville, TN 37996-1301 Telephone: 423-974-5067 / Fax: 423-974-4404 E-Mail: maclennan@cs.utk.edu Homepage: http://www.cs.utk.edu/~mclennan |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-343.ps.Z, *19961210* Client User's Guide to NetSolve Henri Casanova Jack Dongarra y Keith Seymour December 6, 1996 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/Petitet.ps.Z, *19961216* ALGORITHMIC REDISTRIBUTION METHODS FOR BLOCK CYCLIC DECOMPOSITIONS A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Antoine Petitet December 1996 Copyright c 1996 by Antoine Petitet All rights reserved ii To my parents iii Acknowledgments The writer |

| ftp://cs.utk.edu/pub/berry/deer.ps.Z, *19961231* To appear in IEEE Computational Science & Engineering, (1997) Computational Models of White-Tailed Deer in the Florida Everglades* C.A. ABBOTT, M.W. BERRY, J.C. DEMPSEY Department of Computer Science, University of Tennessee, Knoxville TN 37996-1301, fcabbott,berry,dempseyg@cs.utk.edu E.J. COMISKEY, |

| ftp://cs.utk.edu/pub/maclennan/ren.thesis.ps, *19970105* A Natural Language Interface to the Unix Operating System Using Recurrent Neural Networks A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Renwick L. Conrad December 1995 Dedication This thesis is dedicated to my parents, Mr. and Mrs. Lowell Conrad, for without |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-340.ps.Z, *19970106* LAPACK Working Note 116 Parallel Matrix Distributions: Have we been doing it all right Majed Sidani and Bill Harrod Applications Division Cray Research, A Silicon Graphics Company November 14, 1996 * Electronic mail address: sidani@cray.com, harrod@cray.com 1 Parallel Matrix Distributions: Have we been |

| ftp://cs.utk.edu/pub/TechReports/Thesis/koehn.encoding.ps.Z, *19970106* - 1 - Combining Genetic Algorithms and Neural Networks: The Encoding Problem A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Philipp Koehn December 1994 - 2 - Dedication For Claudine Acknowledgment I would like to thank my major professor, Dr. Bruce MacLennan |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-338.ps.Z, *19970106* Reuse Library Interoperability and the World Wide Web Shirley V. Browne University of Tennessee Netlib Development Group browne@cs.utk.edu James W. Moore Chairman of the Reuse Library Interoperability Group (RIG) moorej@acm.org January 2, 1997 1 Prologue Shortly after the collapse of Soviet domination |

| ftp://cs.utk.edu/pub/TechReports/Thesis/ren.thesis.ps.Z, *19970106* A Natural Language Interface to the Unix Operating System Using Recurrent Neural Networks A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Renwick L. Conrad December 1995 Dedication This thesis is dedicated to my parents, Mr. and Mrs. Lowell Conrad, for without |

| ftp://cs.utk.edu/pub/TechReports/Thesis/Mellott.ps.Z, *19970107* A Distributed Implementation of an Individual-Based Predator-Prey Model A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Linda Ellen Mellott May 1997 Acknowledgments I thank my advisor, Dr. Michael Berry, for his encouragement, support, and guidance throughout |

| ftp://cs.utk.edu/pub/maclennan/Elegance.ps, *19970116* Who Cares About Elegance " The Role of Aesthetics in Programming Language Design Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu January 16, 1997 |

| ftp://cs.utk.edu/pub/langston/constructive.ps.Z, *19970117* January 16, 1997 Constructive Algorithms Based on Graph Minors Rajeev Govindany, Michael A. Langstony and Siddharthan Ramachandramurthiz |

| ftp://cs.utk.edu/pub/sost/chapter1.ps, *19970120* DRAFT Chapter 1 Essentials 1.1 ScaLAPACK ScaLAPACK is a library of high performance linear algebra routines for distributed memory message-passing MIMD computers and networks of workstations supporting PVM and/or MPI . It is a continuation of the LAPACK project, which designed and |

| ftp://cs.utk.edu/pub/sost/chapter3.ps, *19970120* DRAFT Chapter 3 Data Distribution 3.1 Introduction ScaLAPACK requires that all global data (vectors or matrices) must be distributed across the processes prior to invoking the ScaLAPACK routines. The storage schemes of global data structures in ScaLAPACK are conceptually the same as for LAPACK. Global |

| ftp://cs.utk.edu/pub/sost/chapter4.ps, *19970120* DRAFT Chapter 4 Documentation and Software Conventions 4.1 Design and Documentation of Argument Lists Like LAPACK, the argument lists of all ScaLAPACK routines conform to a single set of conventions for their design and documentation. Specifications of all ScaLAPACK driver and computational routines are |

| ftp://cs.utk.edu/pub/sost/bibindex.ps, *19970120* DRAFT Bibliography R. Agarwal, F. Gustavson, and M. Zubair, A high performance matrix multiplication algorithm on a distributed-memory parallel computer, using overlapped communication, IBM J. Res. Develop., 38 (1994), pp. 67381. E. Anderson, Z. Bai, C. Bischof, J. Demmel, J. Dongarra, J. Du |

| ftp://cs.utk.edu/pub/sost/chapter7.ps, *19970120* DRAFT Chapter 7 Troubleshooting Successful installation, testing, and use of ScaLAPACK relies heavily on the proper installation of its building blocks (PVM or MPI, BLACS, BLAS, and PBLAS). Test suites are provided for PVM, the BLACS, the BLAS, and the PBLAS. It is highly recommended that each of these |

| ftp://cs.utk.edu/pub/sost/chapter2.ps, *19970120* DRAFT Chapter 2 Contents of ScaLAPACK 2.1 Structure of ScaLAPACK 2.1.1 Levels of Routines The routines in ScaLAPACK are classified into three broad categories: ffl driver routines, each of which solves a complete problem, for example solving a system of linear equations, or computing the eigenvalues of |

| ftp://cs.utk.edu/pub/sost/chapter6.ps, *19970120* DRAFT Chapter 6 Accuracy and Stability Since ScaLAPACK provides the same overall algorithmic approach as LAPACK, the error bounds are the same as for LAPACK. Therefore, this chapter is equivalent to Chapter 4 of the LAPACK Users' Guide, with necessary code segment modifications. In this chapter we |

| ftp://cs.utk.edu/pub/sost/appendices.ps, *19970120* DRAFT Appendix A Index of Driver and Computational Routines Notes 1. This index lists related pairs of real and complex routines together, for example, PSGETRF and PCGETRF. 2. Driver routines are listed in bold type, for example PSGESV and PCGESV. 3. Routines are listed in alphanumeric order of the real |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-345.ps.Z, *19970121* January 16, 1997 Constructive Algorithms Based on Graph Minors Rajeev Govindany, Michael A. Langstony and Siddharthan Ramachandramurthiz |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-344.ps.Z, *19970122* Who Cares About Elegance " The Role of Aesthetics in Programming Language Design Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu January 16, 1997 |

| ftp://cs.utk.edu/pub/sost/appendixj.ps, *19970123* DRAFT Appendix J Issues of Heterogeneous Computing THIS APPENDIX IS IN PREPARATION, WILL BE UPDATED NOV.28! ScaLAPACK runs on heterogeneous, as well as homogeneous, clusters of workstations, and has been thoroughly tested on heterogeneous clusters of workstations using PVM and MPI. Any heterogeneous |

| ftp://cs.utk.edu/pub/maclennan/ut-cs-93-206.ps, *19970124* A Computationally Universal Field Computer That is Purely Linear David H. Wolperty Bruce J. MacLennanz September 14, 1993 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-346.ps.Z, *19970124* Resource Cataloging and Distribution System Keith Moore, Shirley Browne, Jason Cox, and Jonathan Gettler University of Tennessee Technical Report January 1997 |

| ftp://cs.utk.edu/pub/sost/perf.ps, *19970131* DRAFT Chapter 5 Performance of ScaLAPACK Note: this chapter presents some performance figures for ScaLAPACK routines. The figures are provided for illustration only, and should not be regarded as a definitive up-to-date statement of performance. They have been selected from performance figures obtained |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-349.ps.Z, *19970218* Scheduling Block-Cyclic Array Redistribution Fr ed eric Desprez1, Jack Dongarra2;3, Antoine Petitet2, Cyril Randriamaro1 and Yves Robert2 1 LIP, Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France 2 Department of Computer Science, University of Tennessee, Knoxville, TN 37996-1301, USA 3 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-348.ps.Z, *19970218* Computing the Singular Value Decomposition with High Relative Accuracy LAPACK Working Note 119, CS-97-348 James Demmel , Ming Guy, Stanley Eisenstatz, Ivan Slapni<=carx, Kre<=simir Veseli c{, Zlatko Drma<=ck February 11, 1997 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/gshen.ps.Z, *19970227* An Empirical Investigation of Biologically Plausible Hebbian Systems A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville George Shen May 1996 Acknowledgments I would like to thank Dr. Bruce MacLennan for the enthusiasm that he brings to the subject of neural |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-351.ps.Z, *19970228* Interactive and Dynamic Content in Software Repositories Ronald F. Boisvert National Institute of Standards and Technology boisvert@nist.gov Shirley V. Browne University of Tennessee browne@cs.utk.edu Jack J. Dongarra University of Tennessee and Oak Ridge National Laboratory dongarra@cs.utk.edu Eric |

| ftp://cs.utk.edu/pub/plank/classes/cs460/tests/test-95-10-19.ps.Z, *19970306* Midterm Exam CS460: Operating Systems 19 October, 1995 Jim Plank 1 Question 1: 15 Points Part 1 This question concerns interrupt-driven I/O. Specifically, explain what is involved in each of the following: 1. Reading from a byte-serial device like the console. 2. Writing to a byte-serial device like the |

| ftp://cs.utk.edu/pub/plank/classes/cs460/tests/ans-95-10-19.ps.Z, *19970306* Midterm Exam { Answers and grading CS460: Operating Systems 19 October, 1995 Jim Plank Overall stuff So, this test looked ok to me | enough to fill up the time (75 minutes { 15 per question), but not too much. The overall histogram of test scores is as follows: 204060Total: Score0123Number of Students I |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-356.ps.Z, *19970401* Distributed Land-Cover Change Simulation Using PVM and MPI M.W. Berry & K.S. Minser Computer Science Department CS-97-356 February 1997 Distributed Land-Cover Change Simulation Using PVM and MPI M.W. Berry and K.S. Minser Department of Computer Science, University of Tennessee, 107 Ayres Hall, |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m2-2.ps, *19970409* BEBEGHBCBCHBCHBCHBCHBDCHBDCHBCHBFCTab 2ASlot 2BSlot 2CBFCCBF = 60 degrees5 Tetrahedron Module #2 (cont)Open the last two folds up.Fold the top layer backalong line EBFold the remaining two layersinside with an inside fold along line EBNow, inside fold the top layeralong lines GHAnd do the same to the |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m1-1.ps, *19970409* 4.6 inches2.3 inchesABCABC5 Tetrahedron Module #1Start with a 2x1 rectangle whose width is 4.6 inches, and collapse like an accordian:Fold lower left and upper right corners in.Use only the top layer. UnfoldValley fold the entire module along the dotted line and unfold.Bisect angle ABC with a valley |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m2-1.ps, *19970409* 2.5 inchesABCABCDEF60degreesBCEF5 Tetrahedron Module #2Start with a 4x3 rectangle whose width is 2.5 inches, and collapse like an accordian:Fold lower right and upper left corners in. Use only the top layer. Unfold.Valley fold along the dotted line and unfoldFold the top of the upperleft corner (top |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m1-2.ps, *19970409* CDECDEFCFGBCFTab 1ASlot 1BSlot 1CTab 1DGBC = 60 degreesBCF = 97 degrees5 Tetrahedron Module #1 (cont)Fold the upper right corner down again (top layer only), and then fold lineCD (again, top layer only) to line CE (bisecting the 45-degree angle DCE)Mountain fold the entire module along line CF. Unfold |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m3-2.ps, *19970409* ABEFABFGTab 3ATab 3BFAB = GBA = 45 degrees5 Tetrahedron Module #3 (cont)Unfold along lines AF and ABRepeat steps 8-16 on the rightside of the moduleThe finished module:1617 |

| ftp://cs.utk.edu/pub/plank/origami/5tet/m3-1.ps, *19970409* 2.5 inchesABABABABCDABCDABCDEAEABEABF5 Tetrahedron Module #3Start with a 4x3 rectangle whose width is 2.58 inches, and collapse like an accordian:Fold the top layer in half and crease about 1/4 of the way in on each side. Unfold.Then fold the upper left and lower right corners (top layer only) to these |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-354.ps.Z, *19970422* Statistical Performance Modeling: Case Study of the NPB 2.1 Results Erich Strohmaier Computer Science Department, University of Tennessee, Knoxville, TN 37996 UTK-CS-97-354, March 1997 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-347.ps.Z, *19970422* LAPACK Working Note 118 Department of Computer Science Technical Report CS-97-347 The Design and Implementation of the Parallel Out-of-core ScaLAPACK LU, QR and Cholesky Factorization Routines 1 E. F. D'Azevedo Mathematical Sciences Section Oak Ridge National Laboratory P.O. Box 2008, Bldg. 6012 Oak |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-357.ps.Z, *19970422* An Asynchronous Parallel Supernodal Algorithm for Sparse Gaussian Elimination James W. Demmel John R. Gilberty Xiaoye S. Liz February 27, 1997 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-355.ps.Z, *19970422* A Test Matrix Collection for Non-Hermitian Eigenvalue Problems (Release 1.0) Zhaojun Baiy and David Dayz and James Demmelx and Jack Dongarra{ October 6, 1996 1 Introduction The primary purpose of this collection is to provide a testbed for the development of numerical algorithms for solving nonsymmetric |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-342.ps.Z, *19970422* TOP500 Report 1996 Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Science Section Oak Ridge National Laboratory Oak Ridge, TN 37831-6367 dongarra@cs.utk.edu Hans W. Meuer Computing Center University of Mannheim D-68131 Mannheim Germany |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-359.ps.Z, *19970423* Performance Improvements to LAPACK for the Cray Scientific Library Edward Anderson Cray Research 655F Lone Oak Drive Eagan, MN 55121 Mark Fahey y Department of Mathematics University of Kentucky Lexington, KY 40506 April 22, 1997 |

| ftp://cs.utk.edu/pub/TechReports/1996/ut-cs-96-341.ps.Z, *19970425* LAPACK Working Note 117 A Fortran 90 Interface for LAPACK: LAPACK90, version 1.0 L. Susan Blackford Jack J. Dongarray Jeremy Du Crozz Sven Hammarlingx Jerzy Wa sniewski{ December 5, 1996 |

| ftp://cs.utk.edu/pub/plank/classes/cs460/tests/test-95-12-13.ps.Z, *19970501* Final Exam CS460: Operating Systems 13 December, 1995 Jim Plank Do all questions. Note that all questions are not worth the same number of points. In particular: ffl Question 1 is worth 10 points and should take about 10 minutes to answer. ffl Question 2 is worth 25 points and should take about 30 |

| ftp://cs.utk.edu/pub/plank/classes/cs460/tests/ans-95-12-13.ps.Z, *19970501* Final Exam { Questions, Answers and Grading CS460: Operating Systems 13 December, 1995 Jim Plank This was a very challenging exam, and I was very pleased with the results. Most of y'all demonstrated a good capacity to spout both quantitatively and qualitatively about the material, which was one of my |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-360.ps.Z, *19970507* Determining the Idle Time of a Tiling: New Results Fr ed eric Desprez1, Jack Dongarra2;3, Fabrice Rastello1, and Yves Robert2 1 LIP, Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France 2 Department of Computer Science, University of Tennessee, Knoxville, TN 37996-1301, USA 3 Mathematical |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-358.ps.Z, *19970507* Implementation in ScaLAPACK of Divide-and-Conquer Algorithms for Banded and Tridiagonal Linear Systems A. Cleary Department of Computer Science University of Tennessee J. Dongarra Department of Computer Science University of Tennessee Mathematical Sciences Section Oak Ridge National Laboratory |

| ftp://cs.utk.edu/pub/plank/papers/CLIP.ps.Z, *19970516* CLIP: A Checkpointing Tool for Message-Passing Parallel Programs Yuqun Chen, James S. Plank, Kai Li |

| ftp://cs.utk.edu/pub/plank/papers/HPCA97.ps.Z, *19970519* Fault Tolerant Matrix Operations for Networks of Workstations Using Multiple Checkpointing Youngbae Kimy yNERSC, Lawrence Berkeley National Laboratory University of California Berkeley, CA 94720, USA youngbae@nersc.gov James S. Plankz zxDepartment of Computer Science The University of Tennessee |

| ftp://cs.utk.edu/pub/ward/poster.ps, *19970522* ScaLAPACK: A Portable Linear Algebra Library For Distributed Memory Computers L. S. Blackford, J. Choi, A. Cleary, J. Demmel, I. Dhillon, J. Dongarra, S. Hammarling, G. Henry, A. Petitet, K. Stanley, D. Walker, and R. C. Whaley |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-350.ps.Z, *19970527* Tiling with limited resources Pierre-Yves Calland1, Jack Dongarra2;3 and Yves Robert2 1 LIP, Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France 2 Department of Computer Science, University of Tennessee, Knoxville, TN 37996-1301, USA 3 Mathematical Sciences Section, Oak Ridge National |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-362.ps.Z, *19970606* Java Access to Numerical Libraries Henri Casanova Jack Dongarra y David M. Doolin May 31, 1997 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-363.ps.Z, *19970613* A PARALLEL ALGORITHM FOR MESH SMOOTHING LORI FREITAG , MARK JONESy , AND PAUL PLASSMANNz |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-364.ps.Z, *19970613* UNSTRUCTURED MESH COMPUTATIONS ON NETWORKS OF WORKSTATIONS MARK T. JONES AND PAUL E. PLASSMANNy |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-365.ps.Z, *19970619* TOP500 Supercomputer Sites Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Science Section Oak Ridge National Laboratory Oak Ridge, TN 37831-6367 dongarra@cs.utk.edu Hans W. Meuer Computing Center University of Mannheim D-68131 Mannheim |

| ftp://cs.utk.edu/pub/maclennan/digh/cond_thesis.ps.Z, *19970622* The Greek Miracle: An Artificial Life Simulation of the Effects of Literacy on the Dynamics of Communication A Condensed Version of the Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Andrew Douglas Digh December 1994 1 Introduction 1.1 Motivation The original |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-368.ps.Z, *19970707* Sparse Gaussian Elimination on High Performance Computers by Xiaoye S. Li B.S. (Tsinghua University) 1986 M.S., M.A. (Penn State University) 1990 A dissertation submitted in partial satisfaction of the requirements for the degree of Doctor of Philosophy in Computer Science in the GRADUATE DIVISION of |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-371.ps.Z, *19970709* ALGORITHMIC REDISTRIBUTION METHODS FOR BLOCK CYCLIC DECOMPOSITIONS A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Antoine Petitet December 1996 Copyright c 1996 by Antoine Petitet All rights reserved ii To my parents iii Acknowledgments The writer |

| ftp://cs.utk.edu/pub/langston/constraint.ps.Z, *19970710* July 10, 1997 Constraint-Driven Partitioning and Cutting, with Applications to FPGA Design (Extended Abstract) Rajeev Govindan, Michael A. Langston and Barbara C. Plaut Department of Computer Science University of Tennessee Knoxville, TN 37996{1301 frajeev, langston, plautg@cs.utk.edu |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/trees.ps.Z, *19970728* A B A B A B C D E G F C D E F G C D E F G After HookingBefore Hooking After Shrinking |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d30ct.ps.Z, *19970728* Constant Load (2D Torus 30%) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/djsmed.ps.Z, *19970728* 51015202530Processors0110100Seconds2D60%, CM5, DJS3D40%, CM5, DJS2D60%, CM5, RM3D40%, CM5, RM |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d60cr.ps.Z, *19970728* Constant Load (2D Torus 60%) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d40dt.ps.Z, *19970728* Constant Load (3D Torus 40%) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/tertcr.ps.Z, *19970728* Constant Load (Tertiary) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/djsd.ps.Z, *19970728* 50100Processors0110100SecondsRandom 2%, CM5, DJSTertiary, CM5, DJSRandom 2%, CM5, RMTertiary, CM5, RM |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d60dr.ps.Z, *19970728* Constant Load (2D Torus 60%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/spolesen.ps.Z, *19970728* Parallel Computation for Positron Emission Tomography with Reduced Processor Communications A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Soeren Peter Olesen May 1996 Copyright c Soeren Peter Olesen, 1996 All rights reserved ii Acknowledgments This |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d40cr.ps.Z, *19970728* Constant Load (3D Torus 40%) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d20cr.ps.Z, *19970728* Constant Load (3D Torus 20%) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d60dt.ps.Z, *19970728* Constant Load (2D Torus 60%) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d30dr.ps.Z, *19970728* Constant Load (2D Torus 30%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d40dr.ps.Z, *19970728* Constant Load (3D Torus 40%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d20tr.ps.Z, *19970728* Constant Load (3D Torus 20%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1e+00 1.5 2 3 4 5 7 1e+01 1.5 2 3 4 5 7 1e+02 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/tertct.ps.Z, *19970728* Constant Load (Tertiary) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d20ct.ps.Z, *19970728* Constant Load (3D Torus 20%) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/dum.ps.Z, *19970728* 51015202530Processors110100SecondsDJS vs Random Mate on very sparse graphs, CM52D30%, CM5, DJS3D 20%, CM5, DJS2D30%, CM5, RM3D 20%, CM5, RM |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/randdt.ps.Z, *19970728* Constant Load (Random 2%) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/randcr.ps.Z, *19970728* Constant Load (Random 2%) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/mating.ps.Z, *19970728* After MatingBefore Mating A,1 C,1 E,0 B,1 D,0 F,0 A B C F |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d30dt.ps.Z, *19970728* Constant Load (2D Torus 30%) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/svaradhan.ps.Z, *19970728* Estimating the Largest Singular Values/Vectors of Large Sparse Matrices via Modified Moments A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville SOWMINI VARADHAN December 1995 Acknowledgments I thank my dissertation advisor, Dr. Michael Berry, for his |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-372.ps.Z, *19970728* An Overview of Checkpointing in Uniprocessor and Distributed Systems, Focusing on Implementation and Performance James S. Plank Department of Computer Science University of Tennessee Knoxville, TN 37996 plank@cs.utk.edu http://www.cs.utk.edu/~plank Technical Report UT-CS-97-372 Department of Computer |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/segdemo.ps.Z, *19970728* 1 2 3 4 1 1 1 1 2 5 6 7 1 3 4 Segment vector Data vector D |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/tertdt.ps.Z, *19970728* Delta Tree Shrinking Constant Load (Tertiary) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/randct.ps.Z, *19970728* Constant Load (Random 2%) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d30cr.ps.Z, *19970728* Constant Load (2D Torus 30%) 5000 e/p, CM5, RM 10000 e/p, CM5, RM 20000 e/p, CM5, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/plank/papers/CS-97-372.ps.Z, *19970728* An Overview of Checkpointing in Uniprocessor and Distributed Systems, Focusing on Implementation and Performance James S. Plank Department of Computer Science University of Tennessee Knoxville, TN 37996 plank@cs.utk.edu http://www.cs.utk.edu/~plank Technical Report UT-CS-97-372 Department of Computer |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/rmdemo.ps.Z, *19970728* 1 2 Representation of the new graph. 1 1 1 1 1 1 2 3 4 Child segment size 2 2 1 5 2 3 7 6 41 After permute 2 1 2 1 3 1 1 2 After scan 1 4 5 6 After children calculate their own length Unnecessary communication eliminated Now each parent knows the size of its children After Permute 8 4 5 8 1 8 6 After |

| ftp://cs.utk.edu/pub/TechReports/Thesis/kyoungbae.ps.gz, *19970728* Fault Tolerant Matrix Operations for Parallel and Distributed Systems A Dissertation Presented for the Doctor of Philosophy Degree The University of Tennessee, Knoxville Youngbae Kim August 1996 Copyright c Youngbae Kim, 1996 All rights reserved Acknowledgments I dedicate this dissertation to my |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/2d60ct.ps.Z, *19970728* Constant Load (2D Torus 60%) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d20dr.ps.Z, *19970728* Constant Load (3D Torus 20%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/qbrowne.ps.Z, *19970728* Implementing Optical Field Analog Computation with Bacteriorhodopsin: A Feasibility Study A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Quentin A. Browne December 1996 Acknowledgments First and foremost I want to thank GOD for giving me the strength and |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d40ct.ps.Z, *19970728* Constant Load (3D Torus 40%) 5000 e/p, CM5, Tree 10000 e/p, CM5, Tree 20000 e/p, CM5, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/3d20dt.ps.Z, *19970728* Constant Load (3D Torus 20%) 5000 e/p, Delta, Tree 10000 e/p, Delta, Tree 20000 e/p, Delta, Tree Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/rthesis.ps.Z, *19970728* To the Graduate Council: I am submitting herewith a thesis written by Kevin Hart entitled The Performance of PRAM Graph-Connectivity Algorithms on Message-Passing Multiprocessors." I have examined the final copy of this thesis for form and content and recommend that it be accepted in partial |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/tertdr.ps.Z, *19970728* Delta Random Mate Constant Load (Tertiary) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/randdr.ps.Z, *19970728* Constant Load (Random 2%) 5000 e/p, Delta, RM 10000 e/p, Delta, RM 20000 e/p, Delta, RM Seconds Processors 1 1.5 2 3 4 5 7 10 15 20 30 40 50 70 100 0.00 50.00 100.00 150.00 200.00 250.00 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/khart/djssp.ps.Z, *19970728* 51015202530Processors0110100Seconds2D 30%, CM5, DJS3D 20%, CM5, DJS2D 30%, CM5, RM3D 20%, CM5, RM |

| ftp://cs.utk.edu/pub/TechReports/Thesis/digh.ps.Z, *19970729* The Greek Miracle: An Artificial Life Simulation of the Effects of Literacy on the Dynamics of Communication A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Andrew Douglas Digh December 1994 Acknowledgments I wish to thank my advisor, Dr. Bruce J. MacLennan, |

| ftp://cs.utk.edu/pub/TechReports/Thesis/jcrumpton.ps.Z, *19970729* Evolution of Two Symbol Signals by Simulated Organisms A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Joseph J. Crumpton December 1994 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/bhazen.ps.Z, *19970729* A Distributed Implementation of the Land-Use Change Analysis System (LUCAS) Using PVM A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Brett Christopher Hazen August 1995 Acknowledgments I thank my thesis advisor, Dr. Michael W. Berry, for his support and |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-352.ps.Z, *19970730* A Parallel Implementation of the Nonsymmetric QR Algorithm for Distributed Memory Architectures Greg Henry David Watkinsy Jack Dongarraz March 11, 1997 |

| ftp://cs.utk.edu/pub/TechReports/Thesis/jjiang.ps.Z, *19970814* Using Latent Semantic Indexing for Data Mining A Thesis Presented for the Master of Science Degree The University of Tennessee, Knoxville Jingqian Jiang December 1997 Acknowledgments I thank my advisor, Dr. Michael Berry, for his encouragement, support, and guidance throughout this project. I also thank |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-373.ps.gz, *19970818* Tiling for Heterogeneous Computing Platforms Pierre Boulet1, Jack Dongarra2;3, Yves Robert2 and Fr ed eric Vivien1 1 LIP, Ecole Normale Sup erieure de Lyon, 69364 Lyon Cedex 07, France 2 Department of Computer Science, University of Tennessee, Knoxville, TN 37996-1301, USA 3 Mathematical Sciences |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-374.ps.Z, *19970826* On Computing Graph Minor Obstruction Sets Kevin Cattell Department of Computer Science University of Victoria Victoria, B.C. V8W 3P6, Canada kcattell@csr.uvic.ca Michael J. Dinneen Department of Computer Science University of Auckland Auckland, New Zealand mjd@cs.auckland.ac.nz Rodney G. Downey |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-369.ps.Z, *19970919* A New Parallel Matrix Multiplication Algorithm on Distributed-Memory Concurrent Computers Jaeyoung Choi School of Computing Soongsil University 1-1, Sangdo-Dong, Dongjak-Ku Seoul 156-743, KOREA |

| ftp://cs.utk.edu/pub/langston/sets.ps.Z, *19971008* On Computing Graph Minor Obstruction Sets Kevin Cattell Department of Computer Science University of Victoria Victoria, B.C. V8W 3P6, Canada kcattell@csr.uvic.ca Michael J. Dinneen Department of Computer Science University of Auckland Auckland, New Zealand mjd@cs.auckland.ac.nz Rodney G. Downey |

| ftp://cs.utk.edu/pub/berry/cv.ps.gz, *19971027* Vita MICHAEL W. BERRY Work Address Department of Computer Science 107 Ayres Hall University of Tennessee Knoxville, TN 37996-1301 (423) 974-3838 berry@cs.utk.edu http://www.cs.utk.edu/ berry Home Address 1226 Harrington Drive Knoxville, TN 37922 (423) 470-7144 Education UNIVERSITY OF GEORGIA Athens, GA |

| ftp://cs.utk.edu/pub/maclennan/SKI.ps.Z, *19971105* Preliminary Investigation of Random SKI-Combinator Trees Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu October 20, 1997 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-370.ps.Z, *19971114* Preliminary Investigation of Random SKI-Combinator Trees Bruce J. MacLennan Computer Science Department University of Tennessee, Knoxville maclennan@cs.utk.edu October 20, 1997 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-377.ps.Z, *19971118* TOP500 Supercomputer Sites 10th Edition Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Science Section Oak Ridge National Laboratory Oak Ridge, TN 37831-6367 dongarra@cs.utk.edu Hans W. Meuer Computing Center University of Mannheim D-68131 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-378.ps.Z, *19971203* -1- Finding Order in our World: The Primacy of the Concrete in Neural Representations and the Role of Invariance in Substance Reidentification (Extended Version) Commentary on Ruth Garrett Millikan s A Common Structure for Concepts of Individuals, Stuffs, and Real Kinds: More Mama, More Milk and More |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-366.ps.Z, *19980102* Automatically Tuned Linear Algebra Software R. Clint Whaley Jack J. Dongarra Computer Science Department University of Tennessee Knoxville, TN 37996-1301 and Mathematical Sciences Section Oak Ridge National Laboratory Oak Ridge, TN 37831 |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-380.ps.Z, *19980102* Diskless Checkpointing James S. Plank y Kai Li x Michael Puening z y Department of Computer Science University of Tennessee Knoxville, TN 37996 plank@cs.utk.edu x Department of Computer Science Princeton University Princeton, NJ 08544 li@cs.princeton.edu z mpuening@cardinalsolutions.com December 12, |

| ftp://cs.utk.edu/pub/TechReports/1997/ut-cs-97-379.ps.Z, *19980102* Experimental Assessment of Workstation Failures and Their Impact on Checkpointing Systems James S. Plank Wael R. Elwasif Department of Computer Science University of Tennessee Knoxville, TN 37996 @cs.utk.edu December 12, 1997 Technical Report UT-CS-97-379 University of Tennessee Available |