[
https://issues.apache.org/jira/browse/TEZ-1400?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14097621#comment-14097621
]
Bikas Saha commented on TEZ-1400:
---------------------------------
The tez-site change looks good.
For this part. Maybe we should do it in TEZ-744. We should probably not create
a payload if none has been specified. It may confuse some user code.
To maintain compatibility for the ShuffleVertexManager, we should probably
follow up by creating a configuration object for it (see MRInput for an
example) that can be used to configure its parameters via code by
Hive/Pig/others. By default that configuration object can be pick up the params
from client tez-site.xml.
If you agree then please feel free to commit just the tez-site changes in this
jira. I am +1 on it. And fix TEZ-744 separately.
If you think the ShuffleManager config changes make sense then they would be
good to have for the 0.5 release.
{code}
- // Ease of use. If no payload present then give the common configuration
- // TODO TEZ-744 Don't do this - AMConf should not be used to configure
vertexManagers.
+ // Ease of use. If no payload present then give the DAG's configuration
try {
payload = DagTypeConverters.convertToTezUserPayload(
- TezUtils.createUserPayloadFromConf(appContext.getAMConf()));
+
TezUtils.createUserPayloadFromConf(managedVertex.getDAG().getConf()));{code}
> Reducers stuck when enabling auto-reduce parallelism (MRR case)
> ---------------------------------------------------------------
>
> Key: TEZ-1400
> URL: https://issues.apache.org/jira/browse/TEZ-1400
> Project: Apache Tez
> Issue Type: Bug
> Affects Versions: 0.5.0
> Reporter: Rajesh Balamohan
> Assignee: Rajesh Balamohan
> Labels: performance
> Attachments: TEZ-1400.1.patch, TEZ-1400.2.patch, dag.dot
>
>
> In M -> R1 -> R2 case, if R1 is optimized by auto-parallelism R2 gets stuck
> waiting for events.
> e.g
> Map 1: 0/1 Map 2: -/- Map 5: 0/1 Map 6: 0/1 Map 7: 0/1
> Reducer 3: 0/23 Reducer 4: 0/1
> ...
> ...
> Map 1: 1/1 Map 2: 148(+13)/161 Map 5: 1/1 Map 6: 1/1 Map
> 7: 1/1 Reducer 3: 0(+3)/3 Reducer 4: 0(+1)/1 <== Auto reduce
> parallelism kicks in
> ..
> Map 1: 1/1 Map 2: 161/161 Map 5: 1/1 Map 6: 1/1 Map 7: 1/1
> Reducer 3: 3/3 Reducer 4: 0(+1)/1
> Job is stuck waiting for events in Reducer 4.
> [fetcher [Reducer_3] #23]
> org.apache.tez.runtime.library.common.shuffle.impl.ShuffleScheduler: copy(3
> of 23 at 0.02 MB/s) <=== *Waiting for 20 more partitions, even though
> Reducer3 has been optimized to use 3 reducers
--
This message was sent by Atlassian JIRA
(v6.2#6252)
