[
https://issues.apache.org/jira/browse/HADOOP-1448?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel#action_12500452
]
Doug Cutting commented on HADOOP-1448:
--------------------------------------
Other ideas:
1. Limit the maximum replication to something reasonably small.
2. When a file is highly-replicated, sample the replicas rather than sorting
and returning the full list. For example, one could first try to return only
the first-N replicas on the same rack as the client. If that fails, then the
off-rack replicas might be randomly sampled.
Do we think it is reasonable to set replication to numbers proportional to the
cluster size? If so, do we think it is reasonable for namenode requests to
take time proportional to the cluster size? If not, then, in these cases, we
should try to never return or process the full list, but rather only a
fixed-size subset of it.
> Setting the replication factor of a file too high causes namenode cpu overload
> ------------------------------------------------------------------------------
>
> Key: HADOOP-1448
> URL: https://issues.apache.org/jira/browse/HADOOP-1448
> Project: Hadoop
> Issue Type: Bug
> Components: dfs
> Reporter: dhruba borthakur
>
> The replication factor of a file in set to 300 (on a 800 node cluster). Then
> all mappers try to open this file. For every open call that the namenode
> receives from each of these 800 clients, it sorts all the replicas of the
> block(s) based on the distance from the client. This causes CPU usage
> overload on the namenode.
> One proposal is to make the namenode return a non-sorted list of datanodes to
> the client. Information about each replica also contains the rack on which
> that replica resides. The client can look at the replicas to determine if
> there is a copy on the local node. If not, then it can find out if there is a
> replica on the local rack. If not then it can choose a replica at random.
> This proposal is scalable because the sorting and selection of replicas is
> done by the client rather than the Namenode.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.
