NUMA-aware Graph Mining Techniques for Performance and Energy Efficiency
M. Frasca, K. Madduri, and P. Raghavan. Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, page 95:1--95:11. Los Alamitos, CA, USA, IEEE Computer Society Press, (2012)
Abstract
We investigate dynamic methods to improve the power and performance profiles of large irregular applications on modern multi-core systems. In this context, we study a large sparse graph application, Betweenness Centrality, and focus on memory behavior as core count scales. We introduce new techniques to efficiently map the computational demands onto non-uniform memory architectures (NUMA). Our dynamic design adapts to hardware topology and dramatically improves both energy and performance. These gains are more significant at higher core counts.
We implement a scheme for adaptive data layout, which reorganizes the graph after observing parallel access patterns, and a dynamic task scheduler that encourages shared data between neighboring cores. We measure performance and energy consumption on a modern multi-core machine and observe that mean execution time is reduced by 51.2% and energy is reduced by 52.4%.
%0 Conference Paper
%1 frasca12
%A Frasca, Michael R.
%A Madduri, Kamesh
%A Raghavan, Padma
%B Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis
%C Los Alamitos, CA, USA
%D 2012
%I IEEE Computer Society Press
%K graph numa parallel
%P 95:1--95:11
%T NUMA-aware Graph Mining Techniques for Performance and Energy Efficiency
%U http://dl.acm.org/citation.cfm?id=2388996.2389125
%X We investigate dynamic methods to improve the power and performance profiles of large irregular applications on modern multi-core systems. In this context, we study a large sparse graph application, Betweenness Centrality, and focus on memory behavior as core count scales. We introduce new techniques to efficiently map the computational demands onto non-uniform memory architectures (NUMA). Our dynamic design adapts to hardware topology and dramatically improves both energy and performance. These gains are more significant at higher core counts.
We implement a scheme for adaptive data layout, which reorganizes the graph after observing parallel access patterns, and a dynamic task scheduler that encourages shared data between neighboring cores. We measure performance and energy consumption on a modern multi-core machine and observe that mean execution time is reduced by 51.2% and energy is reduced by 52.4%.
%@ 978-1-4673-0804-5
@inproceedings{frasca12,
abstract = {We investigate dynamic methods to improve the power and performance profiles of large irregular applications on modern multi-core systems. In this context, we study a large sparse graph application, Betweenness Centrality, and focus on memory behavior as core count scales. We introduce new techniques to efficiently map the computational demands onto non-uniform memory architectures (NUMA). Our dynamic design adapts to hardware topology and dramatically improves both energy and performance. These gains are more significant at higher core counts.
We implement a scheme for adaptive data layout, which reorganizes the graph after observing parallel access patterns, and a dynamic task scheduler that encourages shared data between neighboring cores. We measure performance and energy consumption on a modern multi-core machine and observe that mean execution time is reduced by 51.2% and energy is reduced by 52.4%.},
acmid = {2389125},
added-at = {2013-02-22T18:58:32.000+0100},
address = {Los Alamitos, CA, USA},
articleno = {95},
author = {Frasca, Michael R. and Madduri, Kamesh and Raghavan, Padma},
biburl = {https://www.bibsonomy.org/bibtex/25ec62d17f2010bf427d35658787e6cfb/ytyoun},
booktitle = {Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis},
interhash = {9dedf389ab71947ab0601f1fb09b2efa},
intrahash = {5ec62d17f2010bf427d35658787e6cfb},
isbn = {978-1-4673-0804-5},
keywords = {graph numa parallel},
location = {Salt Lake City, Utah},
numpages = {11},
pages = {95:1--95:11},
publisher = {IEEE Computer Society Press},
series = {SC '12},
timestamp = {2015-12-12T14:28:45.000+0100},
title = {NUMA-aware Graph Mining Techniques for Performance and Energy Efficiency},
url = {http://dl.acm.org/citation.cfm?id=2388996.2389125},
year = 2012
}