[ https://issues.apache.org/jira/browse/HDFS-9381?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15027655#comment-15027655 ]
Jing Zhao commented on HDFS-9381: --------------------------------- Thanks for working on this, [~umamaheswararao]. I can understand the motivation of this optimization but to me adding a new list in PendingReplication queue may be an overkill: # Currently with a single FSNamesystem lock handling most of the synchronization in both FSNamesystem and BlockManager, compared with the existing huge workload within the lock, the benefit we can gain from this optimization may not be that big. # Originally we have a two-layer replication queue structure: neededReplication <-> pendingReplication. The new list added into the PendingReplication queue actually adds a new layer which complicates the recovery tracking logic. # In case we have a lot of recovery work due to datanode decomm/failure, the same striped block may not be scheduled for recovery in a tight loop. Also considering the striped blocks are more likely to be replicated across multiple racks, I'm not very sure if we will have a lot of such striped blocks with high recovery priority. > When same block came for replication for Striped mode, we can move that block > to PendingReplications > ---------------------------------------------------------------------------------------------------- > > Key: HDFS-9381 > URL: https://issues.apache.org/jira/browse/HDFS-9381 > Project: Hadoop HDFS > Issue Type: Sub-task > Components: erasure-coding, namenode > Affects Versions: 3.0.0 > Reporter: Uma Maheswara Rao G > Assignee: Uma Maheswara Rao G > Attachments: HDFS-9381-02.patch, HDFS-9381-03.patch, > HDFS-9381.00.patch, HDFS-9381.01.patch > > > Currently I noticed that we are just returning null if block already exists > in pendingReplications in replication flow for striped blocks. > {code} > if (block.isStriped()) { > if (pendingNum > 0) { > // Wait the previous recovery to finish. > return null; > } > {code} > Here if we just return null and if neededReplications contains only fewer > blocks(basically by default if less than numliveNodes*2), then same blocks > can be picked again from neededReplications from next loop as we are not > removing element from neededReplications. Since this replication process need > to take fsnamesystmem lock and do, we may spend some time unnecessarily in > every loop. > So my suggestion/improvement is: > Instead of just returning null, how about incrementing pendingReplications > for this block and remove from neededReplications? and also another point to > consider here is, to add into pendingReplications, generally we need target > and it is nothing but to which node we issued replication command. Later when > after replication success and DN reported it, block will be removed from > pendingReplications from NN addBlock. > So since this is newly picked block from neededReplications, we would not > have selected target yet. So which target to be passed to pendingReplications > if we add this block? One Option I am thinking is, how about just passing > srcNode itself as target for this special condition? So, anyway if the block > is really missed, srcNode will not report it. So this block will not be > removed from pending replications, so that when it is timed out, it will be > considered for replication again and that time it will find actual target to > replicate while processing as part of regular replication flow. -- This message was sent by Atlassian JIRA (v6.3.4#6332)