Streaming graph analysis: new models, new architectures

Abstract

Applications in computer network security, social media analysis, health care, and other areas rely on analyzing a changing environment. The data is rich in relationships and lends itself to graph analysis. Traditional static graph analysis cannot keep pace with network security applications analyzing nearly one million events per second. Streaming frameworks like STINGER support ingesting up three million of edge changes per second but there are few streaming analysis kernels that keep up with these rates. We present a new computational model along with promising novel architectures to tackle massive data rates.