Job Scheduling Strategies For Parallel Processing: 10th International Workshop, Jsspp 2004, New York, Ny, Usa, June 13, 2004, Revised Selected Papers (lecture Notes In Computer Science)

This volume contains the papers presented at the 10th Anniversary Workshop on Job Scheduling Strategies for Parallel Processing. The workshop was held in New York City, on June 13, 2004, at Columbia University, in conjunction with the SIGMETRICS 2004 conference. Although it is a workshop, the papers were conference-reviewed, with the full versions being read and evaluated by at least five and usually seven members of the Program Committee. We refer to it as a workshop because of the very fast turnaround time, the intimate nature of the actual presentations, and the ability of the authors to revise their papers after getting feedback from workshop attendees. On the other hand, it was actually a conference in that the papers were accepted solely on their merits as decided upon by the Program Committee. We would like to thank the Program Committee members, Su-Hui Chiang, Walfredo Cirne, Allen Downey, Eitan Frachtenberg, Wolfgang Gentzsch, Allan Gottlieb, Moe Jette, Richard Lagerstrom, Virginia Lo, Reagan Moore, Bill Nitzberg, Mark Squillante, and John Towns, for an excellent job. Thanks are also due to the authors for their submissions, presentations, and final revisions for this volume. Finally, we would like to thank the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), The Hebrew University, and Columbia University for the use of their facilities in the preparation of the workshop and these proceedings. Parallel Job Scheduling, A Status Report / Dror G. Feitelson, Larry Rudolph, Uwe Schwiegelshohn -- Scheduling On The Top 50 Machines / Carsten Ernemann ... [et Al.] -- Parallel Computer Workload Modeling With Markov Chains / Baiyi Song, Carsten Ernemann, Ramin Yahyapour -- Enhancements To The Decision Process Of The Self-tuning Dynp Scheduler / Achim Streit -- Reconfigurable Gang Scheduling Algorithm / Luis Fabricio Wanderley Goes And Carlos Augusto Paiva Da Silva Martins -- Time-critical Scheduling On A Well Utilised Hpc System At Ecmwf Using Loadleveler With Resource Reservation / Graham Holt -- Inferring The Topology And Traffic Load Of Parallel Programs Running In A Virtual Machine Environment / Ashish Gupta And Peter A. Dinda -- Multi-toroidal Interconnects: Using Additional Communication Links To Improve Utilization Of Parallel Computers / Yariv Aridor ... [et Al.] -- Costs And Benefits Of Load Sharing In The Computational Grid / Darin England And Jon B. Weissman -- Workload Characteristics Of A Multi-cluster Supercomputer / Hui Li, David Groep, Lex Wolters -- A Dynamic Co-allocation Service In Multicluster Systems / Jove M.p. Sinaga, Hashim H. Mohamed, Dick H.j. Epema -- Exploiting Replication And Data Reuse To Efficiently Schedule Data-intensive Applications On Grids / Elizeu Santos-neto ... [et Al.] -- Performance Implications Of Failures In Large-scale Cluster Scheduling / Yanyong Zhang ... [et Al.] -- Are User Runtime Estimates Inherently Inaccurate? / Cynthia Bailey Lee ... [et Al.] -- Improving Speedup And Response Times By Replicating Parallel Programs On A Snow / Gaurav D. Ghare And Scott T. Leutenegger -- Lomarc, Lookahead Matchmaking For Multi-resource Coscheduling / Angela C. Sodan And Lei Lan. Dror Feitelson, Larry Rudolph, Uwe Schwiegelshohn (eds.). Includes Bibliographical References And Index. Front Matter....Pages - Parallel Job Scheduling — A Status Report....Pages 1-16 Scheduling on the Top 50 Machines....Pages 17-46 Parallel Computer Workload Modeling with Markov Chains....Pages 47-62 Enhancements to the Decision Process of the Self-Tuning dynP Scheduler....Pages 63-80 Reconfigurable Gang Scheduling Algorithm....Pages 81-101 Time-Critical Scheduling on a Well Utilised HPC System at ECMWF Using Loadleveler with Resource Reservation....Pages 102-124 Inferring the Topology and Traffic Load of Parallel Programs Running in a Virtual Machine Environment....Pages 125-143 Multi-toroidal Interconnects: Using Additional Communication Links to Improve Utilization of Parallel Computers....Pages 144-159 Costs and Benefits of Load Sharing in the Computational Grid....Pages 160-175 Workload Characteristics of a Multi-cluster Supercomputer....Pages 176-193 A Dynamic Co-allocation Service in Multicluster Systems....Pages 194-209 Exploiting Replication and Data Reuse to Efficiently Schedule Data-Intensive Applications on Grids....Pages 210-232 Performance Implications of Failures in Large-Scale Cluster Scheduling....Pages 233-252 Are User Runtime Estimates Inherently Inaccurate?....Pages 253-263 Improving Speedup and Response Times by Replicating Parallel Programs on a SNOW....Pages 264-287 LOMARC — Lookahead Matchmaking for Multi-resource Coscheduling....Pages 288-315 Back Matter....Pages - The popularity of research on the scheduling of parallel jobs demands a periodic review of the status of the field.