A5061702979- Graph Theory and Algorithms
- Distributed systems and fault tolerance
- Cloud Computing and Resource Management
- Distributed and Parallel Computing Systems
University of Minnesota
2017-2018
We present here the results of our investigation a transactional model parallel programming on cluster computing systems. This is specifically targeted for graph applications with goal harnessing unstructured parallelism inherently in many such problems. In this model, tasks vertex-centric computations are executed optimistically as serializable transactions. A key-value based globally shared object store implemented main memory nodes storing data. Task read and modify data distributed...
Summary In dynamically evolving graphs, one may be interested in continuously observing certain properties of the graph. One approach for continuous monitoring is to re‐execute graph analytics program on entire after it updated. However, this can lead high computation cost and latency case large graphs. An alternate approach, which we present paper, execute only initially then perform incremental computations supporting queries as modified. The goal our work develop parallel computing...