Chris Trezzo commented on YARN-1492:

The patch is now +1 overall. Please note that I have broken up this patch into 
smaller patches that are located in each of the sub tasks on this issue. If you 
would like to try out the feature, here are the simple requirements for setting 
up the shared cache:

1. In HDFS, create the shared cache root directory (set to /sharedcache by 

2. In mapred-site.xml add the following parameter:
    A comma delimited list of resource categories to submit to the shared cache.
    The valid categories are: jobjar, libjars, files, archives.
    If "disabled" is specified then the job submission code will not use
    the shared cache.

3. In yarn-site.xml add the following parameter:
    <description>Whether the shared cache is enabled</description>

4. Start the SCM (shared cache manager) using the regular yarn shell scripts.
./yarn-daemon.sh start sharedcachemanager

With this setup all job jars, lib jars, files and archives specified by 
MapReduce jobs will be automatically cached.

> 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: Chris Trezzo
>         Attachments: YARN-1492-all-trunk-v1.patch, 
> YARN-1492-all-trunk-v2.patch, YARN-1492-all-trunk-v3.patch, 
> YARN-1492-all-trunk-v4.patch, YARN-1492-all-trunk-v5.patch, 
> shared_cache_design.pdf, shared_cache_design_v2.pdf, 
> shared_cache_design_v3.pdf, shared_cache_design_v4.pdf, 
> shared_cache_design_v5.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

Reply via email to