Efficient MapReduce algorithms for computing the diameter of very large graphs
Matteo Ceccarello · 2013
Graph data structures are ubiquitous in computer science. Graphs of interest tend to be huge objects, with millions or billions of nodes. To process such graphs, parallel algorithms on distributed architectures are needed. MapReduce is a model of computation developed to process large amounts of data using clusters of commodity machines. One of the most interesting properties of graphs is their diameter. In this thesis we analyze several algorithms to find the diameter of very large graphs