[HDFS-5153] Datanode should send block reports for each storage in a separate message

• Type: Improvement
• Status: Closed
• Priority: Major
• Resolution: Fixed
• Affects Version/s: 3.0.0-alpha1
• Fix Version/s: 2.4.0
• Component/s: datanode
• Target Version/s: 2.4.0
• Hadoop Flags: Reviewed
• Assignee: Arpit Agarwal
• Reporter: Arpit Agarwal
• Created: 30/Aug/13 23:11
• Updated: 12/May/16 18:18
• Resolved: 31/Jan/14 21:31

Description

When the number of blocks on the DataNode grows large we start running into a few issues:

1. Block reports take a long time to process on the NameNode. In testing we have seen that a block report with 6 Million blocks takes close to one second to process on the NameNode. The NameSystem write lock is held during this time.
2. We start hitting the default protobuf message limit of 64MB somewhere around 10 Million blocks. While we can increase the message size limit it already takes over 7 seconds to serialize/unserialize a block report of this size.

HDFS-2832 has introduced the concept of a DataNode as a collection of storages i.e. the NameNode is aware of all the volumes (storage directories) attached to a given DataNode. This makes it easy to split block reports from the DN by sending one report per storage directory to mitigate the above problems.

Issue Links

• breaks: HDFS-6340 DN can't finalize upgrade
• depends upon: HDFS-4988 Datanode must support all the volumes as individual storages
• is duplicated by: HDFS-6756 Default ipc.maximum.data.length should be increased to 128MB from 64MB
• is related to: HDFS-7579 Improve log reporting during block report rpc failure