Abstract
Graph workloads exhibit highly irregular memory access patterns, resulting in poor cache utilization. By modifying the layout of the stored graph prior to processing, cache utilization can be enhanced. Two factors need to be considered for modifying the layout of the graph. First, the nature of computation in vertex-centric algorithms suggests that vertex neighbours are visited in succession throughout processing. Second, the degree distribution of vertices in real-world networks exhibits power-law distribution, implying that a few vertices are responsible for the majority of the connections. As a result, such nodes can be clustered together to improve temporal and spatial locality. In this paper, we propose Community Aware Graph Reordering (CAR), which leverages both these aspects to enhance the performance of graph applications when compared to existing reordering strategies. While previous state-of-the-art reordering techniques with comparable reordering overheads, such as Hub Cluster, DBG, and Sorder, deliver speedup of 9%, 11%, and 17%, respectively, CAR provides a speedup of 20%.
S. Singhania and N. Sharma—Both authors have contributed equally
C. K. Jha—The work was carried out when Chandan Kumar Jha was at CADSL, IIT Bombay.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Cuzzocrea, A., et al.: Big graph analytics: the state of the art and future research agenda. In: Proceedings of the 17th International Workshop on Data Warehousing and OLAP, pp. 99–101 (2014)
Beamer, S., et al.: Direction-optimizing breadth-first search. In: SC 2012: Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis, pp. 1–10. IEEE (2012)
Shiloach, Y., et al.: An O(logn) parallel connectivity algorithm. J. Algorithms 3(1), 57–67 (1982)
Beamer, S., et al.: The gap benchmark suite. arXiv preprint arXiv:1508.03619 (2015)
Balaji, V., et al.: 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), pp. 203–214. IEEE (2018)
Faldu, P., et al.: A closer look at lightweight graph reordering. In: 2019 IEEE International Symposium on Workload Characterization (IISWC), pp. 1–13. IEEE (2019)
Wei, H., et al.: Speedup graph processing by graph ordering. In: Proceedings of the 2016 International Conference on Management of Data, pp. 1813–1828 (2016)
Leskovec, J., et al.: SNAP Datasets: Stanford large network dataset collection, June 2014. http://snap.stanford.edu/data
Huang, B., et al.: Structure preserved graph reordering for fast graph processing without the pain. In: 2020 IEEE 22nd International Conference on High Performance Computing and Communications; IEEE 18th International Conference on Smart City; IEEE 6th International Conference on Data Science and Systems (HPCC/SmartCity/DSS), pp. 44–51. IEEE (2020)
Mellor-Crummey, J., et al.: Improving memory hierarchy performance for irregular applications using data and computation reorderings. Int. J. Parallel Programm. 29(3), 217–247 (2001)
Yu, X. et al.: IMP: indirect memory prefetcher. In: Proceedings of the 48th International Symposium on Microarchitecture, pp. 178–190 (2015)
Faldu, P., et al.: Domain-specialized cache management for graph analytics. In: 2020 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 234–248. IEEE (2020)
Basak, A., et al.: Analysis and optimization of the memory hierarchy for graph processing workloads. In: 2019 IEEE International Symposium on High Performance Computer Architecture (HPCA), pp. 373–386. IEEE (2019)
Balaji, V., et al.: P-OPT: practical optimal cache replacement for graph analytics. In: 2021 IEEE International Symposium on High-Performance Computer Architecture (HPCA), pp. 668–681. IEEE (2021)
Sharma, N., et al.: Data-aware cache management for graph analytics. In: 2022 Design, Automation and Test in Europe Conference and Exhibition (DATE), pp. 843–848. IEEE (2022)
Intel VTune Profiler (2021). https://software.intel.com/content/www/us/en/develop/tools/oneapi/components/vtune-profiler.html#gs.34axdf
Wu, C.-J., et al.: SHiP: signature-based hit predictor for high performance caching. In: Proceedings of the 44th Annual IEEE/ACM International Symposium on Microarchitecture, pp. 430–441 (2011)
Jain, A., et al.: Back to the future: leveraging Belady’s algorithm for improved cache replacement. In: 2016 ACM/IEEE 43rd Annual International Symposium on Computer Architecture (ISCA), pp. 78–89. IEEE (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Singhania, S., Sharma, N., Venkitaraman, V., Jha, C.K. (2022). CAR: Community Aware Graph Reordering for Efficient Cache Utilization in Graph Analytics. In: Shah, A.P., Dasgupta, S., Darji, A., Tudu, J. (eds) VLSI Design and Test. VDAT 2022. Communications in Computer and Information Science, vol 1687. Springer, Cham. https://doi.org/10.1007/978-3-031-21514-8_37
Download citation
DOI: https://doi.org/10.1007/978-3-031-21514-8_37
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-21513-1
Online ISBN: 978-3-031-21514-8
eBook Packages: Computer ScienceComputer Science (R0)