YuAng Chen
ABSTRACT
The intrinsic irregular data structure of graphs often causes poor cache utilization thus deteriorates the performance of graph analytics. Prior works have designed a variety of graph reordering methods to improve cache efficiency. However, little insight has been provided into the issue of workload imbalance for multicore systems. In this work, we identify that a major factor affecting the performance is the unevenly distributed computation load amongst cores. To cope with this problem, we propose cache-aware reordering (Corder), a lightweight reordering algorithm that facilitates workload balance as well as cache optimization. Comprehensive performance evaluation of Corder is conducted on various graph applications and datasets. We observe that Corder yields speedup of up to 2.59× (on average 1.47×) over original graphs.
- V. Balaji and B. Lucia. 2018. When is Graph Reordering an Optimization? Studying the Effect of Lightweight Graph Reordering Across Applications and Input Graphs. In 2018 IEEE International Symposium on Workload Characterization (IISWC). 203--214.Google Scholar
- Priyank Faldu, Jeff Diamond, and Boris Grot. 2019. A closer look at lightweight graph reordering. In 2019 IEEE International Symposium on Workload Characterization (IISWC). IEEE, 1--13.Google ScholarCross Ref
- Kartik Lakhotia, Rajgopal Kannan, Sourav Pati, and Viktor Prasanna. 2020. GPOP: A Scalable Cache-and Memory-efficient Framework for Graph Processing over Parts. ACM Transactions on Parallel Computing (TOPC) 7, 1 (2020), 1--24.Google ScholarDigital Library
- Julian Shun and Guy E Blelloch. 2013. Ligra: a lightweight graph processing framework for shared memory. In Proceedings of the 18th ACM SIGPLAN symposium on Principles and practice of parallel programming. 135--146.Google ScholarDigital Library
- Yunming Zhang, Vladimir Kiriansky, Charith Mendis, Saman Amarasinghe, and Matei Zaharia. 2017. Making caches work for graph analytics. In 2017 IEEE International Conference on Big Data (Big Data). IEEE, 293--302.Google ScholarCross Ref
Index Terms
- Corder: cache-aware reordering for optimizing graph analytics
Recommendations
Minnow: Lightweight Offload Engines for Worklist Management and Worklist-Directed Prefetching
ASPLOS '18: Proceedings of the Twenty-Third International Conference on Architectural Support for Programming Languages and Operating SystemsThe importance of irregular applications such as graph analytics is rapidly growing with the rise of Big Data. However, parallel graph workloads tend to perform poorly on general-purpose chip multiprocessors (CMPs) due to poor cache locality, low ...
Minnow: Lightweight Offload Engines for Worklist Management and Worklist-Directed Prefetching
ASPLOS '18The importance of irregular applications such as graph analytics is rapidly growing with the rise of Big Data. However, parallel graph workloads tend to perform poorly on general-purpose chip multiprocessors (CMPs) due to poor cache locality, low ...
Architectural exploration of last-level caches targeting homogeneous multicore systems
SBCCI '16: Proceedings of the 29th Symposium on Integrated Circuits and Systems Design: Chip on the MountainsThe Last-Level Cache (LLC) influences the overall system performance and power dissipation in multicore systems significantly. This paper evaluates five LLC architectures targeting execution time, dynamic and static power dissipation, and area ...
Comments