High Performance Scientific and Engineering Computing

by Tianruo Yang, Laurence; Pan, Yi

• ISBN13: 9781402075803
• ISBN10: 1402075804
• Format: Hardcover
• Copyright: 2003-12-01
• Publisher: Kluwer Academic Pub

Summary

The field of high performance computing achieved prominence through advances in electronic and integrated technologies beginning in the 1940s. Current times are very exciting and the years to come will witness a proliferation of the use of parallel and distributed systems. The scientific and engineering application domains have a key role in shaping future research and development activities in academia and industry, especially when the solution of large and complex problems must cope with harder and harder timing. High Performance Scientific And Engineering Computing: Hardware/Software Support contains selected chapters on hardware/software support for high performance scientific and engineering computing from prestigious workshops in the fields such as PACT-SHPSEC, IPDPS-PDSECA and ICPP-HPSECA. This edited volume is basically divided into six main sections which include invited material from prominent researchers around the world. We believe all of these contributed chapters and topics not only provide novel ideas, new results and state-of-the-art techniques in this field, but also stimulate the future research activities in the area of high performance computing for science and engineering applications. High Performance Scientific And Engineering Computing: Hardware/Software Support is designed for a professional audience, composed of researchers and practitioners in industry. This book is also suitable as a secondary text for graduate-level students in computer science and engineering.

Table of Contents

Preface	xi
Part I Compilation and Architectural Support
1 Multithreaded Parallelism with OpenMP	3	(12)
Raimi A. Rufai, Muslim Bozyigit, Jarallah S. AlGhamdi, and Moataz Ahmed
1 Introduction	3	(2)
2 Background	5	(3)
3 Performance Studies	8	(2)
4 Conclusion and Further Work	10	(2)
Appendix: Source Listing	12	(3)
2 Linear Data Distribution based on Index Analysis	15	(16)
Minyi Guo
1 Introduction	15	(1)
2 Overview of Data Distribution	16	(1)
3 Motivation	17	(1)
4 Linear Data Distribution	18	(6)
5 Index Conversion and Iteration Space Conversion	24	(2)
6 Experimental Results for Linear Distribution	26	(2)
7 Summary	28	(3)
3 Quantification of Memory Communication	31	(14)
Surendra Byna, Kirk W. Cameron and Xian-He Sun
1 Introduction	31	(2)
2 Related Work	33	(1)
3 Motivating Example	34	(2)
4 Experimental Details	36	(1)
5 Quantifying Communication Cost	37	(2)
6 Identifying Memory Communication Buffers	39	(3)
7 Conclusions and Future Work	42	(3)
4 The Need for Adaptive Dynamic Thread Scheduling	45	(18)
Chulho Shin, Seong- Won Lee and Jean-Luc Gaudiot
1 Introduction	46	(1)
2 Related Work	47	(1)
3 Adaptive Dynamic Thread Scheduling with the Detector Thread	48	(4)
4 Methodology	52	(3)
5 Experimental Results	55	(3)
6 Summary and Conclusion	58	(5)
Part II Numerical Computation
5 Performance Analysis of a BiCGSTAB Solver for Multiple-Marine-Propeller Simulation with Several MPI Libraries and Platforms	63	(16)
Penfei Liu and Kun Li
1 Introduction	63	(2)
2 Computational Configurations	65	(2)
3 Using the MPI Libraries	67	(4)
4 Results and Discussion	71	(6)
5 Conclusions	77	(2)
6 A New BLAS-3 Based Parallel Algorithm for Computing the Eigenvectors of Real Symmetric Matrices	79	(16)
Yusaku Yamamoto and Mitsuyoshi Igai and Ken Naono
1 Introduction	80	(1)
2 Review of the Conventional Inverse Iteration Method	81	(2)
3 The Householder Inverse Iteration Method	83	(3)
4 The Blocked Algorithm	86	(1)
5 Numerical Results	87	(4)
6 Conclusion	91	(4)
Part III Load Balancing
7 Load Balancing in SPMD Applications: Concepts and Experiments	95	(14)
A. Plastmo, V. Thomé, D. Vianna, R. Costa, and O.T. da Silveira Filho
1 Introduction	95	(1)
2 Classification of Load Balancing Algorithms	96	(3)
3 Load Imbajancing Index	99	(1)
4 Load Balancing in a Scientific SPMD Application	99	(7)
5 Concluding Remarks	106	(3)
8 Dynamic Load Balance Strategy: Application to Nonlinear Optics	109	(18)
A. Bourgeade and B. Nkonga
1 Introduction	109	(1)
2 Nonlinear Maxwell-Lorentz Equations	110	(2)
3 Numerical Approach	112	(3)
4 Message Passing Approach	115
5 Applications	111	(13)
6 Conclusion	124	(3)
9 Message-Passing Parallel Adaptive Quantum Trajectory Method	127	(16)
R.L. Cariño, I. Banicescu and R.K. Vadapalli, C.A. Weatherford and J. Zhu
1 Introduction	128	(1)
2 The Hydrodynamic Formulation of Quantum Mechanics	129	(1)
3 Simulation of Wavepacket Dynamics	130	(2)
4 The Moving Weighted Least Squares Algorithm	132	(1)
5 Loop Scheduling	132	(3)
6 Experiments	135	(2)
7 Concluding Remarks	137	(6)
Part IV Performance Evaluation
10 Verifying Large-Scale System Performance during Installation using Modelling	143	(14)
Darren J. Kerbyson, Adolfy Hoisie and Harvey J. Wasserman
1 Introduction	143	(1)
2 Performance Modelling	144	(3)
3 The Alpha-Server ES45 Supercomputing System	147	(1)
4 The Application and the Model	148	(1)
5 Use of the SAGE Model to Validate System Performance	148	(7)
6 Summary	155	(2)
11 Mixed Level Modelling and Simulation of Large Scale HW/SW Systems	157	(12)
Murali K. Nethi and James H. Aylor
1 Introduction	157	(1)
2 Performance Estimation Techniques	158	(2)
3 Our Modelling Approach	160	(2)
4 Mixed Level System Simulation Environment	162	(1)
5 Preliminary Results	163	(1)
6 Conclusions and Future Work	164	(5)
Part V Grid Computing
12 Engineering an Autonomic Partitioning Framework for Grid-based SAMB Applications	169	(20)
S. Chandra, X. Li, and M. Passhar
1 Introduction	169	(2)
2 Problem Description	171	(3)
3 ARMaDA: An Autonomic SAMB Partitioning Framework	174	(6)
4 An Experimental Evaluation of Autonomic Partitioning Policies	180	(5)
5 Summary and Conclusions	185	(4)
13 A Resource Discovery Service For a Mobile Agents Based Grid Infrastructure	189	(12)
R. Aversa, B. Di Martino, N. Mazzocca and S. Venticinque
1 Introduction	189	(1)
2 The Grid And Web Services Models	190	(2)
3 A Magda Overview	192	(2)
4 The Discovery And Invocation Service	194	(3)
5 Related Work	197	(4)
14 xDGDL: Towards an XML-based DataGrid Description Language	201	(16)
Erich Schikuta
1 Introduction	201	(1)
2 A Novel File Hierarchy	202	(3)
3 xDGDL - the XML Data Grid Description Language	205	(4)
4 A Practical xDGDL Example	209	(2)
5 An Application of xDGDL	211	(1)
6 Conclusions and Future Work	212	(5)
Part VI Scientific and Engineering Applications
15 A Large-scale MD Simulation for Formation Process of Carbon Cluster on a Parallel Computer	217	(14)
Ryoko Hayashi, Kenji Tanaka, Susumu Horiguchi and Yasuaki Hiwatari
1 Introduction	218	(1)
2 Basic information	218	(7)
3 Parallelization Process	225	(2)
4 Parallelization Results	227	(2)
5 Approximation of Execution Times and Outstanding Problems	229	(1)
6 Conclusion	230	(1)
16 Comparing Various Parallelizing Approaches for Tribology Simulations	231	(22)
V. Chaudhary, W.L. Hase, H. Jiang, L. Sun, and D. Thaker
1 Introduction	231	(2)
2 Parallelization Approaches	233	(5)
3 Molecular Dynamics	238	(4)
4 Experiments and Analysis	242	(2)
5 Conclusion and Future Work	244	(5)
Appendix: A. Sequential Code	249	(2)
Appendix: B. OpenMP Code	251	(1)
Appendix: C. MPI Code	251	(1)
Appendix: D. Strings DSM Code	252	(1)
17 High Performance Air Pollution Simulation on Shared Memory Systems	253	(14)
Maria I. Martin, Marta Parada and Ramón Doallo
1 Introduction	253	(1)
2 Air Pollution Modelling	254	(1)
3 Sequential Profile of the STEM-II Program	255	(3)
4 Optimization of the Sequential Code	258	(3)
5 Parallel Code Using OpenMP	261	(6)
18 An Ant Colony Optimization Based Routing Algorithm in Mobile Ad hoc Networks and its Parallel Implementation	267	(18)
Parimala Thulasiraman, Ruppa K. Thulasiram, and Mohammad T. Islam
1 Introduction	268	(1)
2 Background	269	(4)
3 Parallel ACO Routing Algorithm	273	(4)
4 Experimental Results	277	(3)
5 Conclusions	280	(5)
19 Parallelizing Serializable Transactions Based on Transaction Classification in Real-time Database Systems	285	(14)
Subhash Bhalla and Masaki Hasegawa
1 Introduction	285	(1)
2 Nature of Delays	286	(2)
3 The Proposed Model	288	(2)
4 Definitions for Real-time Database System	290	(1)
5 Ordering of Transactions in a Distributed System	291	(1)
6 An Algorithm to Construct LAG	292	(2)
7 Performance Evaluation	294	(3)
8 Summary and Conclusions	297	(2)
20 Adaptive Selection of Materialized Queries in a Mediator for the Integration of Distributed Information Resources	299	(16)
Kil Hong Joo and Won Suk Lee
1 Introduction	299	(2)
2 Related Work	301	(2)
3 Decayed Usage Patterns	303	(2)
4 Implementation Plans	305	(5)
5 Experiments and Analysis of Results	310	(2)
6 Conclusion	312	(3)
Index	315