pero on anything

November 30th, 2009

Improve performance on small hadoop clusters

Hadoop is designed to run on huge clusters containing several hundred machines. But some people just don’t need such a big cluster and are able to use the benefits of HDFS and MapReduce on a smaller scale. We managed to improve performance of our 10-node-test-cluster by almost 100% by adjusting the heartbeat intervals. Namenode and [...]