An optimal design method for parallel computing is presented based on algorithm graph.
研究基于算法图的并行计算优化设计方法。
Based on Spark distributed graph computing model, a large scale and complex network community parallel discovery algorithm DBCS is proposed.
基于Spark分布式图计算模型,提出大规模复杂网络社区并行发现算法DBCS。
Experiments show that the new optimized partition algorithm can improve the parallel simulation performance compared with the ones partitioned by traditional graph partition algorithms.
实验表明,该优化划分方法相对于传统的图划分算法能够有效地提高模拟性能。
应用推荐