Learn More
Graph partitioning and repartitioning have been widely used by scientists to parallelize compute- and dataintensive simulations. However, existing graph (re)partitioning algorithms usually assume homogeneous communication costs among partitions, which contradicts the increasing heterogeneity in inter-core communication in modern parallel architectures and(More)
With the explosion of large, dynamic graph datasets from various fields, graph partitioning and repartitioning are becoming more and more critical to the performance of many graph-based Big Data applications , such as social analysis, web search, and recommender systems. However, well-studied graph (re)partitioners usually assume a homogeneous and(More)
Graph partitioning is an essential preprocessing step in distributed graph computation and scientific simulations. Existing well-studied graph partitioners are designed for static graphs, but real-world graphs, such as social networks and Web networks, keep changing dynamically. In fact, the communication and computation patterns of some graph algorithms(More)
The increasing popularity and ubiquity of various large graph datasets has caused renewed interest for graph partitioning. Existing graph partitioners either scale poorly against large graphs or disregard the impact of the underlying hardware topology. A few solutions have shown that the nonuniform network communication costs may affect the performance(More)
Developed an efficient benchmark to measure the relative network communication costs among machines. Developed three architecture-aware graph (re)partitioners using MPI for vertex-centric distributed graph computation, achieving up to 12x speedups on three classic graph workloads: BFS, SSSP and PageRank. Developed a skew-resistant graph partitioner for(More)
  • 1