作者: Anand Tripathi , Vinit Padhye , Tara Sasank Sunkara
关键词: Computer science 、 Power graph analysis 、 Graph (abstract data type) 、 Distributed computing 、 Graph 、 Maximum flow problem 、 Theoretical computer science 、 Cloud computing 、 Spanning tree 、 Optimistic concurrency control 、 Graph coloring 、 Transactional memory
摘要: Beehive is a parallel programming framework designed for cluster-based computing environments in cloud data centers. It specifically targeted graph analysis problems. The provides the abstraction of key-value based global object storage, which maintained memory cluster nodes. Its computation model on optimistic concurrency control executing concurrent tasks as atomic transactions harnessing amorphous parallelism We describe here architecture and abstractions provided by this framework, present performance several problems such maximum flow, minimum weight spanning tree, coloring, PageRank algorithm.