hadoop-hdfs-issues mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Tsz Wo Nicholas Sze (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (HDFS-5168) BlockPlacementPolicy does not work for cross node group dependencies
Date Tue, 29 Apr 2014 03:56:16 GMT

    [ https://issues.apache.org/jira/browse/HDFS-5168?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13983960#comment-13983960

Tsz Wo Nicholas Sze commented on HDFS-5168:

DNSToSwitchMapping is a \@Public \@Evolving interface so that we have to change it in a compatible
manner (otherwise, we cannot commit this to branch-2) .  We should avoid adding the new getDependency(..)
method to it.  How about we add another interface class, say DNSToSwitchMappingWithDependency,
and keep DNSToSwitchMapping unchanged?

More details:
- DNSToSwitchMappingWithDependency extends DNSToSwitchMapping and adds the new getDependency(..)
- ScriptBasedMappingWithDependency extends ScriptBasedMapping and RawScriptBasedMappingWithDependency
extends RawScriptBasedMapping; change ScriptBasedMapping and RawScriptBasedMapping to allow
- Add dependency cache support to ScriptBasedMappingWithDependency.
- DatanodeManager checks if dnsToSwitchMapping instanceof DNSToSwitchMappingWithDependency.
 If yes, cast the object and get dependencies; otherwise, use empty list.
- CachedDNSToSwitchMapping and TableMapping remains unchanged.

> BlockPlacementPolicy does not work for cross node group dependencies
> --------------------------------------------------------------------
>                 Key: HDFS-5168
>                 URL: https://issues.apache.org/jira/browse/HDFS-5168
>             Project: Hadoop HDFS
>          Issue Type: Improvement
>          Components: namenode
>            Reporter: Nikola Vujic
>            Assignee: Nikola Vujic
>            Priority: Critical
>         Attachments: HDFS-5168.patch, HDFS-5168.patch, HDFS-5168.patch, HDFS-5168.patch,
HDFS-5168.patch, HDFS-5168.patch
> Block placement policies do not work for cross rack/node group dependencies. In reality
this is needed when compute servers and storage fall in two independent fault domains, then
both BlockPlacementPolicyDefault and BlockPlacementPolicyWithNodeGroup are not able to provide
proper block placement.
> Let's suppose that we have Hadoop cluster with one rack with two servers, and we run
2 VMs per server. Node group topology for this cluster would be:
>  server1-vm1 -> /d1/r1/n1
>  server1-vm2 -> /d1/r1/n1
>  server2-vm1 -> /d1/r1/n2
>  server2-vm2 -> /d1/r1/n2
> This is working fine as long as server and storage fall into the same fault domain but
if storage is in a different fault domain from the server, we will not be able to handle that.
For example, if storage of server1-vm1 is in the same fault domain as storage of server2-vm1,
then we must not place two replicas on these two nodes although they are in different node
> Two possible approaches:
> - One approach would be to define cross rack/node group dependencies and to use them
when excluding nodes from the search space. This looks as the cleanest way to fix this as
it requires minor changes in the BlockPlacementPolicy classes.
> - Other approach would be to allow nodes to fall in more than one node group. When we
chose a node to hold a replica we have to exclude from the search space all nodes from the
node groups where the chosen node belongs. This approach may require major changes in the

This message was sent by Atlassian JIRA

View raw message