[HDFS-17458] Remove unnecessary BP lock in ReplicaMap

• Type: Improvement
• Status: Open
• Priority: Major
• Resolution: Unresolved
• Affects Version/s: 3.4.0
• Component/s: datanode
• Labels: pull-request-available
• Assignee: farmmamba
• Reporter: farmmamba
• Created: 09/Apr/24 05:59
• Updated: 29/Apr/24 04:03

Description

In HDFS-16429 we make LightWeightResizableGSet to be thread safe, and in HDFS-16511  we change some methods in ReplicaMap to acquire read lock instead of acquiring write lock.

This PR try to remove unnecessary Block_Pool read lock further.

Recently, I performed stress tests on datanodes to measure their read/write operations/second.

Before we removing some lock,  it can only achieve ~2K write ops. After optimizing, it can achieve more than 5K write ops.

Issue Links

• links to: GitHub Pull Request #6717