Characteristics of a Large Shared Memory Production Workload

This paper characterizes the production workload that highly utilizes the NCSA Origin 2000. The characterization includes the distributions of job interarrival time, requested number of processors, requested memory, requested runtime, actual runtime as a fraction of requested runtime, and the ratio of memory usage to memory request. Conditional distributions are defined as needed for generating a synthetic workload with the same characteristics, including the key correlations observed among the job parameters. Characteristics of the O2K workload that differ from previously reported production workload characteristics are also noted.

[1]  Dror G. Feitelson,et al.  Memory Usage in the LANL CM-5 Workload , 1997, JSSPP.

[2]  Dror G. Feitelson,et al.  Utilization and Predictability in Scheduling the IBM SP2 with Backfilling , 1998, Proceedings of the First Merged International Parallel Processing Symposium and Symposium on Parallel and Distributed Processing.

[3]  Timothy J. O'Donnell,et al.  Analysis of the early workload on the Cornell Theory Center IBM SP2 , 1996, SIGMETRICS '96.

[4]  Dror G. Feitelson,et al.  Packing Schemes for Gang Scheduling , 1996, JSSPP.

[5]  Allen B. Downey.  Predicting queue times on space-sharing parallel computers , 1997, Proceedings 11th International Parallel Processing Symposium.

[6]  Thomas R. Gross,et al.  Impact of Job Mix on Optimizations for Space Sharing Schedulers , 1996, Proceedings of the 1996 ACM/IEEE Conference on Supercomputing.

[7]  Reagan Moore,et al.  A Batch Scheduler for the Intel Paragon MPP System with a Non-contiguous Node Allocation Algorithm , 1996, JSSPP.

[8]  Arnold O. Allen,et al.  Probability, statistics and queueing theory - with computer science applications (2. ed.) , 1981, Int. CMG Conference.

[9]  Mor Harchol-Balter.  Task assignment with unknown duration , 2002, JACM.

[10]  Mor Harchol-Balter,et al.  Exploiting process lifetime distributions for dynamic load balancing , 1995, SIGMETRICS.

[11]  Dror G. Feitelson,et al.  Job Characteristics of a Production Parallel Scientivic Workload on the NASA Ames iPSC/860 , 1995, JSSPP.

[12]  Richard Gibbons,et al.  A Historical Application Profiler for Use by Parallel Schedulers , 1997, JSSPP.

[13]  Allen B. Downey,et al.  The elusive goal of workload characterization , 1999, PERV.

[14]  Fang Wang,et al.  Modeling of Workload in MPPs , 1997, JSSPP.

[15]  Mark S. Squillante,et al.  The impact of job memory requirements on gang-scheduling performance , 1999, PERV.

[16]  Mark S. Squillante,et al.  The impact of job arrival patterns on parallel scheduling , 1999, PERV.

[17]  A. W. Kemp,et al.  Univariate Discrete Distributions , 1993 .

[18]  Steven Hotovy,et al.  Workload Evolution on the Cornell Theory Center IBM SP2 , 1996, JSSPP.

[19]  Mary K. Vernon,et al.  Production job scheduling for parallel shared memory systems , 2001, Proceedings 15th International Parallel and Distributed Processing Symposium. IPDPS 2001.

[20]  Bill Nitzberg,et al.  A comparison of workload traces from two production parallel machines , 1996, Proceedings of 6th Symposium on the Frontiers of Massively Parallel Computation (Frontiers '96).