Ghent University Academic Bibliography

Advanced

Using cycle stacks to understand scaling bottlenecks in multi-threaded workloads

Wim Heirman UGent, Trevor Carlson UGent, Shuai Che, Kevin Skadron and Lieven Eeckhout UGent (2011) International Symposium on Workload Characterization Proceedings. p.38-49
abstract
This paper proposes a methodology for analyzing parallel performance by building cycle stacks. A cycle stack quantifies where the cycles have gone, and provides hints towards optimization opportunities. We make the case that this is particularly interesting for analyzing parallel performance: understanding how cycle components scale with increasing core counts and/or input data set sizes leads to insight with respect to scaling bottlenecks due to synchronization, load imbalance, poor memory performance, etc. We present several case studies illustrating the use of cycle stacks. As a subsequent step, we further extend the methodology to analyze sets of parallel workloads using statistical data analysis, and perform a workload characterization to understand behavioral differences across benchmark suites. We analyze the SPLASH-2, PARSEC and Rodinia benchmark suites and conclude that the three benchmark suites cover similar areas in the workload space. However, scaling behavior of these benchmarks towards larger input sets and/or higher core counts is highly dependent on the benchmark, the way in which the inputs have been scaled, and on the machine configuration.
Please use this url to cite or link to this publication:
author
organization
year
type
conference
publication status
published
subject
in
International Symposium on Workload Characterization Proceedings
issue title
2011 IEEE International symposium on workload characterization (IISWC)
pages
38 - 49
publisher
IEEE
place of publication
New York, NY, USA
conference name
2011 IEEE International symposium on Workload Characterization (IISWC 2011)
conference location
Austin, TX, USA
conference start
2011-11-06
conference end
2011-11-08
Web of Science type
Proceedings Paper
Web of Science id
000299350700004
ISBN
9781457720635
9781457720642
9781457720628
DOI
10.1109/IISWC.2011.6114195
language
English
UGent publication?
yes
classification
P1
copyright statement
I have transferred the copyright for this publication to the publisher
id
1959548
handle
http://hdl.handle.net/1854/LU-1959548
date created
2011-12-03 14:53:46
date last changed
2013-06-21 12:56:16
@inproceedings{1959548,
  abstract     = {This paper proposes a methodology for analyzing parallel performance by building cycle stacks. A cycle stack quantifies where the cycles have gone, and provides hints towards optimization opportunities. We make the case that this is particularly interesting for analyzing parallel performance: understanding how cycle components scale with increasing core counts and/or input data set sizes leads to insight with respect to scaling bottlenecks due to synchronization, load imbalance, poor memory performance, etc.
We present several case studies illustrating the use of cycle stacks. As a subsequent step, we further extend the methodology to analyze sets of parallel workloads using statistical data analysis, and perform a workload characterization to understand behavioral differences across benchmark suites. We analyze the SPLASH-2, PARSEC and Rodinia benchmark suites and conclude that the three benchmark suites cover similar areas in the workload space. However, scaling behavior of these benchmarks towards larger input sets and/or higher core counts is highly dependent on the benchmark, the way in which the inputs have been scaled, and on the machine configuration.},
  author       = {Heirman, Wim and Carlson, Trevor and Che, Shuai and Skadron, Kevin and Eeckhout, Lieven},
  booktitle    = {International Symposium on Workload Characterization Proceedings},
  isbn         = {9781457720635},
  language     = {eng},
  location     = {Austin, TX, USA},
  pages        = {38--49},
  publisher    = {IEEE},
  title        = {Using cycle stacks to understand scaling bottlenecks in multi-threaded workloads},
  url          = {http://dx.doi.org/10.1109/IISWC.2011.6114195},
  year         = {2011},
}

Chicago
Heirman, Wim, Trevor Carlson, Shuai Che, Kevin Skadron, and Lieven Eeckhout. 2011. “Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-threaded Workloads.” In International Symposium on Workload Characterization Proceedings, 38–49. New York, NY, USA: IEEE.
APA
Heirman, W., Carlson, T., Che, S., Skadron, K., & Eeckhout, L. (2011). Using cycle stacks to understand scaling bottlenecks in multi-threaded workloads. International Symposium on Workload Characterization Proceedings (pp. 38–49). Presented at the 2011 IEEE International symposium on Workload Characterization (IISWC 2011), New York, NY, USA: IEEE.
Vancouver
1.
Heirman W, Carlson T, Che S, Skadron K, Eeckhout L. Using cycle stacks to understand scaling bottlenecks in multi-threaded workloads. International Symposium on Workload Characterization Proceedings. New York, NY, USA: IEEE; 2011. p. 38–49.
MLA
Heirman, Wim, Trevor Carlson, Shuai Che, et al. “Using Cycle Stacks to Understand Scaling Bottlenecks in Multi-threaded Workloads.” International Symposium on Workload Characterization Proceedings. New York, NY, USA: IEEE, 2011. 38–49. Print.