Cluster Rebalancing

Kevin Feasel



Peter Coates discusses cluster rebalancing in Hadoop:

After adding new racks to our 70 node cluster, we noticed that it was taking several hours per terabyte to rebalance the nodes. You can copy a terabyte of data across a 10GbE network in under half an hour with SCP, so why should HDFS take several hours?

It didn’t take long to discover the cause—the configuration parameterdfs.datanode.balance.bandwidthPerSecond controls how much bandwidth each node is allowed to use for rebalancing, and it defaults to a conservative value of 10Mb/sec/node, which is 1.25MB/sec. If you have 70 nodes (the number we started with before adding new ones), that’s 87.5MB/second. One terabyte, i.e., a million MB, divided 87.5MB/sec, equals 11,428 sec, or 3.17 hours per TB. The more nodes in the original cluster, the faster it will write.

On the development side, “it’ll automatically rebalance without us having to worry” is a great thing.  On the administrative side, we’re paid to worry about these things…

Related Posts

Flink and Stateful Streaming

Himanshu Gupta explains some of the benefits Apache Flink offers for stateful streaming applicatons: The 2 main types of stream processing done are:1. Stateless: Where every event is handled completely independent from the preceding events.2. Stateful: Where a “state” is shared between events and therefore past events can influence the way current events are processed. […]

Read More

Performance Testing Aiven Kafka

Heikki Nousiainen tests the Aiven platform’s Kafka implementation on different cloud providers at different service levels: We used a single topic for our write operations with a partition count set to either 3 or 6, depending on the number of brokers in each test cluster. As the test clusters were regular Aiven services, the partitions […]

Read More


July 2016
« Jun Aug »