I'm using giraph-1.3.0.-SNAPSHOT and hadoop-2.8.4 in a EC2 cluster composed of 5 nodes (everyone has 32 cpus and 60 GBs ram).
If I give small input to my algorithm implemented in giraph, It properly works. When I give a large input (like https://snap.stanford.edu/data/web-NotreDame.html) I obtains this error and my giraph run fails: WARN server.NIOServerCnxn: caught end of stream exception EndOfStreamException: Unable to read additional data from client sessionid likely client has closed socket I think it is related to zookeeper but I'm not sure. -- Francesco Sclano