- Graph Connect is an industry conference that focuses on graph databases, primarily Neo4j.
- BigGraphs 2015 The Second International Workshop on High Performance Big Graph Data Management, Analysis, and Mining (BigGraphs 2015) To be held in conjunction with IEEE BigData 2015 Oct 29--Nov 1, 2015, Santa Clara, CA, USA.
Distributed graph processing platforms:
- GraphX is an extension to Spark to provide Scala-based APIs for graph computation. See this paper for more information.
- PowerGraph is the core of GraphLab, which has an interesting graph partitioning component and focuses on machine learning algorithms.
- Giraph is the Apache counterpart of Pregel, the graph processing system used in Google.
Shared-memory graph processing platforms:
- Galois is a lightweight graph infrastructure, which supports speculative parallelization, asynchronous computation, and deterministic parallelism.
- GraphChi can handle big graphs that do not fit into main memory.
- Poymer has a special data representation to match NUMA machines.
- Powerlyra supports differential computation for big graphs.
- X-stream is the first edge-centric graph processing system for large graphs that do not fit into main memory.
Graph processing on GPUs:
All of the following systems leverage Nvidia GPUs to accelerate graph processing. They all assume the input graphs can fit into device memory.
- The data sets here are used for graph partitioning competition.
- These data sets are mainly used for web algorithm research.
- These data sets have many different types of graphs in terms of size and topology.
- This website lists machines that can process graphs efficiently. It provides a BFS benchmark and a graph generator.