Collaborative local triangles counting in a graph stream by using node resources

Sever based analysis for local triangles counting cannot handle large scale graph accurately. For accuracy it needs a large amount of memory and it is almost impossible for the server to prepare all. So till now for efficiency the server uses a small amount of memory and finds out the approximate value of local triangles. In this paper we propose the method in which we use edge device resources in order to count local triangles. In this method edge devices collaborate with the server for counting local triangles. It can improve sever-based analysis effectively. Furthermore it can find out the exact number of local triangles.