[
https://issues.apache.org/jira/browse/HDFS-8999?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15074407#comment-15074407
]
Jing Zhao commented on HDFS-8999:
---------------------------------
bq. Here you are optimizing completeBlock(), not IBR, which I don't know how
substantial it is, if at all.
bq. I suggest that we change the code to allow closing a file even if the last
block is in COMMITTED state in this JIRA and then change DN to send IBRs in
batches in another JIRA.
I think Nicholas's comment has summarized our final goal: to decrease the total
number of IBR. Currently writing a block with 3 replicas can easily generate >6
RPCs, which greatly limits the scalability of HDFS handling small files. We
should explore if we can batch IBR and let DN send them periodically. The first
step will be breaking the dependency between {{completeFile}} and IBR.
> Namenode need not wait for {{blockReceived}} for the last block before
> completing a file.
> -----------------------------------------------------------------------------------------
>
> Key: HDFS-8999
> URL: https://issues.apache.org/jira/browse/HDFS-8999
> Project: Hadoop HDFS
> Issue Type: Bug
> Components: namenode
> Reporter: Jitendra Nath Pandey
> Assignee: Tsz Wo Nicholas Sze
> Attachments: h8999_20151228.patch
>
>
> This comes out of a discussion in HDFS-8763. Pasting [~jingzhao]'s comment
> from the jira:
> {quote}
> ...whether we need to let NameNode wait for all the block_received msgs to
> announce the replica is safe. Looking into the code, now we have
> # NameNode knows the DataNodes involved when initially setting up the
> writing pipeline
> # If any DataNode fails during the writing, client bumps the GS and
> finally reports all the DataNodes included in the new pipeline to NameNode
> through the updatePipeline RPC.
> # When the client received the ack for the last packet of the block (and
> before the client tries to close the file on NameNode), the replica has been
> finalized in all the DataNodes.
> Then in this case, when NameNode receives the close request from the client,
> the NameNode already knows the latest replicas for the block. Currently the
> checkReplication call only counts in all the replicas that NN has already
> received the block_received msg, but based on the above #2 and #3, it may be
> safe to also count in all the replicas in the
> BlockUnderConstructionFeature#replicas?
> {quote}
--
This message was sent by Atlassian JIRA
(v6.3.4#6332)
