[
https://issues.apache.org/jira/browse/YARN-1492?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13859231#comment-13859231
]
Karthik Kambatla commented on YARN-1492:
----------------------------------------
Comments on the design:
# In the client protocol, if a cleaner instance (or run) starts after R2 and
before R2', the client wouldn't know of this cleaner's existence.
# Dangling cleaner locks: Using ZK here would probably make it easier to handle
these dangling locks. If the Cleaner crashes, the corresponding connection to
ZK is severed, and all locks are automatically cleaned up (if using ephemeral
nodes). As others have mentioned earlier, I think it is okay to assume one ZK
quorum running. For instance, RM HA requires this.
# We should probably mandate running CleanerService if shared-cache is enabled,
and should run as part of the RM and periodically.
> truly shared cache for jars (jobjar/libjar)
> -------------------------------------------
>
> Key: YARN-1492
> URL: https://issues.apache.org/jira/browse/YARN-1492
> Project: Hadoop YARN
> Issue Type: New Feature
> Affects Versions: 2.0.4-alpha
> Reporter: Sangjin Lee
> Assignee: Sangjin Lee
> Attachments: shared_cache_design.pdf, shared_cache_design_v2.pdf,
> shared_cache_design_v3.pdf, shared_cache_design_v4.pdf
>
>
> Currently there is the distributed cache that enables you to cache jars and
> files so that attempts from the same job can reuse them. However, sharing is
> limited with the distributed cache because it is normally on a per-job basis.
> On a large cluster, sometimes copying of jobjars and libjars becomes so
> prevalent that it consumes a large portion of the network bandwidth, not to
> speak of defeating the purpose of "bringing compute to where data is". This
> is wasteful because in most cases code doesn't change much across many jobs.
> I'd like to propose and discuss feasibility of introducing a truly shared
> cache so that multiple jobs from multiple users can share and cache jars.
> This JIRA is to open the discussion.
--
This message was sent by Atlassian JIRA
(v6.1.5#6160)