zhtttylz commented on PR #5759: URL: https://github.com/apache/hadoop/pull/5759#issuecomment-1610928161
> > Thanks to involve me here. Skim the PR (seem very common logic), I am wonder if this issue is only related with EC feature? or will it also impact on replication scenario? > > @Hexiaoqiao The problem solved by this PR is only related to `BlockPlacementPolicyRackFaultTolerant`. `BlockPlacementPolicyRackFaultTolerant` has implemented a placement policy that ensures each replica to be placed on a different rack. By default, EC files use this policy. Replicated files can also use it if really desired. @zhangshuyan0 's explanation is excellent,I have benefited immensely, and I would like to provide a brief supplement. `BlockPlacementPolicyRackFaultTolerant` enables the distribution of blocks across multiple racks. It is the default configuration for the `dfs.block.placement.ec.classname` Erasure Coding block placement policy. Additionally, we can also be enabled for BlockPlacementPolicyRackFaultTolerant by configuring `dfs.block.replicator.classname` to enable the BlockPlacementPolicyRackFaultTolerant for replicas. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
