A practical method for estimating performance degradation on multicore processors, and its application to HPC workloads

@article{Dwyer2012APM,
  title={A practical method for estimating performance degradation on multicore processors, and its application to HPC workloads},
  author={Tyler Dwyer and Alexandra Fedorova and Sergey Blagodurov and Mark Roth and Fabien Gaud and Jian Pei},
  journal={2012 International Conference for High Performance Computing, Networking, Storage and Analysis},
  year={2012},
  pages={1-11}
}
When multiple threads or processes run on a multi-core CPU they compete for shared resources, such as caches and memory controllers, and can suffer performance degradation as high as 200%. We design and evaluate a new machine learning model that estimates this degradation online, on previously unseen workloads, and without perturbing the execution. Our motivation is to help data center and HPC cluster operators effectively use workload consolidation. Data center consolidation is about placing… CONTINUE READING

Citations

Publications citing this paper.
Showing 1-10 of 41 extracted citations

Contention aware frequency scaling on CMPs with guaranteed quality of service

2014 Design, Automation & Test in Europe Conference & Exhibition (DATE) • 2014
View 4 Excerpts
Highly Influenced

Estimating Effective Slowdown of Tasks in Energy-Aware Clouds

2014 IEEE International Symposium on Parallel and Distributed Processing with Applications • 2014
View 5 Excerpts
Highly Influenced

Characterization and modeling of PIDX parallel I/O for performance optimization

2013 SC - International Conference for High Performance Computing, Networking, Storage and Analysis (SC) • 2013
View 5 Excerpts
Highly Influenced

Practical Iterative Optimization for the Data Center

TACO • 2015
View 3 Excerpts
Highly Influenced

A performance enforcing mechanism for energy- and failure-aware cloud systems

International Green Computing Conference • 2014
View 3 Excerpts
Highly Influenced

References

Publications referenced by this paper.
Showing 1-3 of 3 references

Mechanistic-empirical processor performance modeling for constructing CPI stacks on real hardware

(IEEE ISPASS) IEEE INTERNATIONAL SYMPOSIUM ON PERFORMANCE ANALYSIS OF SYSTEMS AND SOFTWARE • 2011
View 3 Excerpts
Highly Influenced

Maui Scheduler Open Cluster Software

Sourceforge
http://mauischeduler.sourceforge.net/.
View 3 Excerpts
Highly Influenced

Similar Papers

Loading similar papers…