With the growth of computer network, graphs are ubiquitous nowadays, social networks, papers citing network, world wide web to a few. On the other hand, with the drastic drop of storage cost and the emergence of large social network companies like Facebook and Linkedin, graphs are now of a unprecedented size with billions of vertexes and edges. Mining such large dataset may help us gain lots of useful information and leads to interesting applications anomaly detection, social network analysis and so on. There are also lots of famous graph mining algorithms like PageRank, Random Walk with Restart, Belief Propagation who are aiming to find interesting patterns in graphs. However, most of these algorithms assume the graph fit in memory, which is apparently no longer suitable for the giant graphs today