The Community for Technology Leaders
2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT) (2012)
Minneapolis, MN, USA
Sept. 19, 2012 to Sept. 23, 2012
ISBN: 978-1-5090-6609-4
pp: 457-458
David Eklov , Uppsala University, Department of Information Technology, Sweden
Nikos Nikoleris , Uppsala University, Department of Information Technology, Sweden
David Black-Schaffer , Uppsala University, Department of Information Technology, Sweden
Erik Hagersten , Uppsala University, Department of Information Technology, Sweden
ABSTRACT
Applications that are co-scheduled on a multi-core compete for shared resources, such as cache capacity and memory bandwidth. The performance degradation resulting from this contention can be substantial, which makes it important to effectively manage these shared resources. This, however, requires quantitative insight into how applications are impacted by such contention. In this paper we present a quantitative method to measure applications' sensitivities to different degrees of contention for off-chip memory bandwidth on real hardware. We then use the data captured with our profiling method to estimate the throughput of a set of co-running instances of a single threaded application.
INDEX TERMS
Bandwidth, Throughput, Sensitivity, Correlation, Benchmark testing, Memory management, Information technology,Performance Prediction, Memory contention
CITATION
David Eklov, Nikos Nikoleris, David Black-Schaffer, Erik Hagersten, "Bandwidth Bandit: Quantitative characterization of memory contention", 2012 21st International Conference on Parallel Architectures and Compilation Techniques (PACT), vol. 00, no. , pp. 457-458, 2012, doi:
99 ms
(Ver 3.3 (11022016))