Advanced search
1 file | 1.74 MB Add to list

Improving IBM POWER8 performance through symbiotic job scheduling

Author
Organization
Abstract
Symbiotic job scheduling, i.e., scheduling applications that co-run well together on a core, can have a considerable impact on the performance of processors with simultaneous multithreading (SMT) cores. SMTcores share most of their microarchitectural components among the co-running applications, which causes performance interference between them. Therefore, scheduling applications with complementary resource requirements on the same core can greatly improve the throughput of the system. This paper enhances symbiotic job scheduling for the IBM POWER8 processor. We leverage the existing cycle accounting mechanism to build an interference model that predicts symbiosis between applications. The proposed models achieve higher accuracy than previous models by predicting job symbiosis from throttled CPI stacks, i.e., CPI stacks of the applications when running in the same SMT mode to consider the statically partitioned resources, but without interference from other applications. The symbiotic scheduler uses these interference models to decide, at run-time, which applications should run on the same core or on separate cores. We prototype the symbiotic scheduler as a user-level scheduler in the Linux operating system and evaluate it on an IBM POWER8 server running multiprogram workloads. The symbiotic job scheduler significantly improves performance compared to both an agnostic random scheduler and the default Linux scheduler. Across all evaluated workloads in SMT4 mode, throughput improves by 12.4 and 5.1 percent on average over the random and Linux schedulers, respectively.
Keywords
SMT PROCESSORS, Symbiotic job scheduling, performance estimation, interference model, IBM POWER8, simultaneous multithreading (SMT)

Downloads

  • (...).pdf
    • full text
    • |
    • UGent only
    • |
    • PDF
    • |
    • 1.74 MB

Citation

Please use this url to cite or link to this publication:

MLA
Feliu, Josue et al. “Improving IBM POWER8 Performance Through Symbiotic Job Scheduling.” IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS 28.10 (2017): 2838–2851. Print.
APA
Feliu, J., Eyerman, S., Sahuquillo, J., Petit, S., & Eeckhout, L. (2017). Improving IBM POWER8 performance through symbiotic job scheduling. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, 28(10), 2838–2851.
Chicago author-date
Feliu, Josue, Stijn Eyerman, Julio Sahuquillo, Salvador Petit, and Lieven Eeckhout. 2017. “Improving IBM POWER8 Performance Through Symbiotic Job Scheduling.” Ieee Transactions on Parallel and Distributed Systems 28 (10): 2838–2851.
Chicago author-date (all authors)
Feliu, Josue, Stijn Eyerman, Julio Sahuquillo, Salvador Petit, and Lieven Eeckhout. 2017. “Improving IBM POWER8 Performance Through Symbiotic Job Scheduling.” Ieee Transactions on Parallel and Distributed Systems 28 (10): 2838–2851.
Vancouver
1.
Feliu J, Eyerman S, Sahuquillo J, Petit S, Eeckhout L. Improving IBM POWER8 performance through symbiotic job scheduling. IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS. Los alamitos: Ieee Computer Soc; 2017;28(10):2838–51.
IEEE
[1]
J. Feliu, S. Eyerman, J. Sahuquillo, S. Petit, and L. Eeckhout, “Improving IBM POWER8 performance through symbiotic job scheduling,” IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, vol. 28, no. 10, pp. 2838–2851, 2017.
@article{8542509,
  abstract     = {Symbiotic job scheduling, i.e., scheduling applications that co-run well together on a core, can have a considerable impact on the performance of processors with simultaneous multithreading (SMT) cores. SMTcores share most of their microarchitectural components among the co-running applications, which causes performance interference between them. Therefore, scheduling applications with complementary resource requirements on the same core can greatly improve the throughput of the system. This paper enhances symbiotic job scheduling for the IBM POWER8 processor. We leverage the existing cycle accounting mechanism to build an interference model that predicts symbiosis between applications. The proposed models achieve higher accuracy than previous models by predicting job symbiosis from throttled CPI stacks, i.e., CPI stacks of the applications when running in the same SMT mode to consider the statically partitioned resources, but without interference from other applications. The symbiotic scheduler uses these interference models to decide, at run-time, which applications should run on the same core or on separate cores. We prototype the symbiotic scheduler as a user-level scheduler in the Linux operating system and evaluate it on an IBM POWER8 server running multiprogram workloads. The symbiotic job scheduler significantly improves performance compared to both an agnostic random scheduler and the default Linux scheduler. Across all evaluated workloads in SMT4 mode, throughput improves by 12.4 and 5.1 percent on average over the random and Linux schedulers, respectively.},
  author       = {Feliu, Josue and Eyerman, Stijn and Sahuquillo, Julio and Petit, Salvador and Eeckhout, Lieven},
  issn         = {1045-9219},
  journal      = {IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS},
  keywords     = {SMT PROCESSORS,Symbiotic job scheduling,performance estimation,interference model,IBM POWER8,simultaneous multithreading (SMT)},
  language     = {eng},
  number       = {10},
  pages        = {2838--2851},
  publisher    = {Ieee Computer Soc},
  title        = {Improving IBM POWER8 performance through symbiotic job scheduling},
  url          = {http://dx.doi.org/10.1109/TPDS.2017.2691708},
  volume       = {28},
  year         = {2017},
}

Altmetric
View in Altmetric
Web of Science
Times cited: