Hi, I took a quick shot at Semi Clustering algorithm in the Google Pregel Paper using Hama. I have to yet test and clean the code. Here is the code
https://github.com/praveensripati/hama-algorithms/tree/master/src/org/apache/hama/examples/SemiClusteting I need some help in this regard. I have defined the SemiClusterVertex as below public class SemiClusterVertex extends Vertex<Text, IntWritable, MapWritable> { } According to the algorithm each of the vertex holds all the semi clusters that hold the vertex, so I choose a MapWritable with Key containing an IntWritable and the value the SemiCluster object. For some reason o.a.h.io.MapWritable doesn't have a toString() method. Should I change MapWritable with a new toString() and build a Hadoop jar? Praveen,
