Graph edge partition models have recently become an appealing alternative to graph vertex partition models for distributed computing due to both their flexibility in balancing loads and their performance in reducing communication cost.In this paper, we propose a simple yet effective graph edge partitioning algorithm. In practice, our algorithm provides good partition quality while maintaining low partition overhead. It also outperforms similar state-of-the-art edge partition approaches, especially for power-law graphs. In theory, previous work showed that an approximation guarantee of O(dmax√(log n log k)) apply to the graphs with m=Ω(k2) edges (n is the number of vertices, and k is the number of partitions). We further rigorously proved that this approximation guarantee hold for all graphs.We also demonstrate the applicability of the proposed edge partition algorithm in real parallel computing systems. We draw our example from GPU program locality enhancement and demonstrate that the graph edge partition model does not only apply to distributed computing with many computer nodes, but also to parallel computing in a single computer node with a many-core processor.
References
2017
POMACS
A Simple Yet Effective Balanced Edge Partition Model for Parallel Computing
Lingda Li, Robel Geda, Ari B. Hayes, and
4 more authors
Graph edge partition models have recently become an appealing alternative to graph vertex partition models for distributed computing due to their flexibility in balancing loads and their performance in reducing communication cost [6, 16]. In this paper, we propose a simple yet effective graph edge partitioning algorithm. In practice, our algorithm provides good partition quality (and better than similar state-of-the-art edge partition approaches, at least for power-law graphs) while maintaining low partition overhead. In theory, previous work [6] showed that an approximation guarantee of O(dmax√ log n log k) apply to the graphs with m=Ω(k2) edges (k is the number of partitions). We further rigorously proved that this approximation guarantee hold for all graphs.We show how our edge partition model can be applied to parallel computing. We draw our example from GPU program locality enhancement and demonstrate that the graph edge partition model does not only apply to distributed computing with many computer nodes, but also to parallel computing in a single computer node with a many-core processor.
@article{lietalpomacs17,author={Li, Lingda and Geda, Robel and Hayes, Ari B. and Chen, Yanhao and Chaudhari, Pranav and Zhang, Eddy Z. and Szegedy, Mario},title={A Simple Yet Effective Balanced Edge Partition Model for Parallel Computing},year={2017},issue_date={June 2017},publisher={Association for Computing Machinery},address={New York, NY, USA},volume={1},number={1},doi={10.1145/3084451},journal={Proc. ACM Meas. Anal. Comput. Syst.},month=jun,articleno={14},numpages={21},keywords={gpu, data sharing, program locality, graph model, edge partition}}
SIGMETRICS
A Simple Yet Effective Balanced Edge Partition Model for Parallel Computing
Lingda Li, Robel Geda, Ari B. Hayes, and
4 more authors
In Proceedings of the 2017 ACM SIGMETRICS / International Conference on Measurement and Modeling of Computer Systems, Jun 2017
Graph edge partition models have recently become an appealing alternative to graph vertex partition models for distributed computing due to both their flexibility in balancing loads and their performance in reducing communication cost.In this paper, we propose a simple yet effective graph edge partitioning algorithm. In practice, our algorithm provides good partition quality while maintaining low partition overhead. It also outperforms similar state-of-the-art edge partition approaches, especially for power-law graphs. In theory, previous work showed that an approximation guarantee of O(dmax√(log n log k)) apply to the graphs with m=Ω(k2) edges (n is the number of vertices, and k is the number of partitions). We further rigorously proved that this approximation guarantee hold for all graphs.We also demonstrate the applicability of the proposed edge partition algorithm in real parallel computing systems. We draw our example from GPU program locality enhancement and demonstrate that the graph edge partition model does not only apply to distributed computing with many computer nodes, but also to parallel computing in a single computer node with a many-core processor.
@inproceedings{lietalsigmetrics17,author={Li, Lingda and Geda, Robel and Hayes, Ari B. and Chen, Yanhao and Chaudhari, Pranav and Zhang, Eddy Z. and Szegedy, Mario},title={A Simple Yet Effective Balanced Edge Partition Model for Parallel Computing},year={2017},isbn={9781450350327},publisher={Association for Computing Machinery},address={New York, NY, USA},doi={10.1145/3078505.3078520},booktitle={Proceedings of the 2017 ACM SIGMETRICS / International Conference on Measurement and Modeling of Computer Systems},pages={6},numpages={1},keywords={edge partition, GPU, program locality, graph model, data sharing},location={Urbana-Champaign, Illinois, USA},series={SIGMETRICS '17 Abstracts}}