ASAP: fast, approximate graph pattern mining at scale

摘要: While there has been a tremendous interest in processing data that an underlying graph structure, existing distributed systems take several minutes or even hours to mine simple patterns on graphs. This paper presents ASAP, fast, approximate computation engine for pattern mining. ASAP leverages state-of-the-art results approximation theory, and extends it general settings. To enable the users navigate tradeoff between result accuracy latency, we propose novel approach build Error-Latency Profile (ELP) given computation. We have implemented general-purpose dataflow platform evaluated extensively patterns. Our experimental show outperforms exact mining solutions by up 77×. Further, can scale graphs with billions of edges without need large clusters.