[
https://issues.apache.org/jira/browse/TEZ-2363?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14514719#comment-14514719
]
Gopal V commented on TEZ-2363:
--
[~hitesh]: the failed test was a network timeout.
Counters:
[
https://issues.apache.org/jira/browse/TEZ-2363?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2363:
-
Fix Version/s: 0.7.0
Counters: off by 1 error for REDUCE_INPUT_GROUPS counter
[
https://issues.apache.org/jira/browse/TEZ-2407?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14527144#comment-14527144
]
Gopal V edited comment on TEZ-2407 at 5/4/15 7:49 PM:
--
No, the issue is
[
https://issues.apache.org/jira/browse/TEZ-2407?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14527144#comment-14527144
]
Gopal V commented on TEZ-2407:
--
No, the issue is that {{DataInputBuffer::getLength()}} has bad
[
https://issues.apache.org/jira/browse/TEZ-145?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14491654#comment-14491654
]
Gopal V commented on TEZ-145:
-
bq. Not sure why pipelining is required for this? Essentially we
[
https://issues.apache.org/jira/browse/TEZ-1993?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14481929#comment-14481929
]
Gopal V commented on TEZ-1993:
--
Committed to trunk, thanks [~bikassaha].
This is going to be
[
https://issues.apache.org/jira/browse/TEZ-1993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-1993:
-
Attachment: TEZ-1993.3.patch
Rebase to trunk for commit.
Implement a pluggable InputSizeEstimator for grouping
[
https://issues.apache.org/jira/browse/TEZ-1993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-1993:
-
Attachment: (was: TEZ-1993.3.patch)
Implement a pluggable InputSizeEstimator for grouping fairly
[
https://issues.apache.org/jira/browse/TEZ-1993?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-1993:
-
Attachment: TEZ-1993.3.patch
Implement a pluggable InputSizeEstimator for grouping fairly
[
https://issues.apache.org/jira/browse/TEZ-2251?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14391159#comment-14391159
]
Gopal V commented on TEZ-2251:
--
Is vertex auto-parallelism kicking in when it receives 25% of
[
https://issues.apache.org/jira/browse/TEZ-145?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14490535#comment-14490535
]
Gopal V commented on TEZ-145:
-
Combiner have only made sense in case of 3/4.
#3 is the true use
[
https://issues.apache.org/jira/browse/TEZ-145?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14486728#comment-14486728
]
Gopal V commented on TEZ-145:
-
[~ozawa]: Got some time to test this out.
The current
[
https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14574146#comment-14574146
]
Gopal V commented on TEZ-2244:
--
The patch looks good, except for a commented block in the
[
https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14589280#comment-14589280
]
Gopal V commented on TEZ-2244:
--
[~rajesh.balamohan]: this patch has some more issues for what
[
https://issues.apache.org/jira/browse/TEZ-2505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14568557#comment-14568557
]
Gopal V commented on TEZ-2505:
--
[~hitesh]: to match the original mrx code, I suggest passing in
[
https://issues.apache.org/jira/browse/TEZ-2505?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14568303#comment-14568303
]
Gopal V commented on TEZ-2505:
--
I took that on as a known limitation after looking at the
[
https://issues.apache.org/jira/browse/TEZ-625?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-625:
Summary: DAG/Vertex names should be restricted to using [A-Za-z0-9 _] and
limited to a defined character limit (was:
[
https://issues.apache.org/jira/browse/TEZ-1997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14694131#comment-14694131
]
Gopal V commented on TEZ-1997:
--
bq. I don't think access to totalKeys is thread safe in this
[
https://issues.apache.org/jira/browse/TEZ-1997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14694186#comment-14694186
]
Gopal V edited comment on TEZ-1997 at 8/12/15 9:36 PM:
---
bq. Read
[
https://issues.apache.org/jira/browse/TEZ-1997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14694186#comment-14694186
]
Gopal V commented on TEZ-1997:
--
bq. Read protected by nothing.
Yes, because it is read when a
[
https://issues.apache.org/jira/browse/TEZ-2728?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2728:
-
Attachment: hive.log.gz
Wrap IPC connection Exception as SessionNotRunning - RM crash
Gopal V created TEZ-2728:
Summary: Wrap IPC connection Exception as SessionNotRunning - RM
crash
Key: TEZ-2728
URL: https://issues.apache.org/jira/browse/TEZ-2728
Project: Apache Tez
Issue Type:
[
https://issues.apache.org/jira/browse/TEZ-2728?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2728:
-
Description:
Crashing the RM when a query session is open and restarting it does not result
in a recoverable state
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14707099#comment-14707099
]
Gopal V commented on TEZ-2731:
--
[~rajesh.balamohan]: confirmed that the perf bottleneck exists
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2731:
-
Issue Type: Sub-task (was: Improvement)
Parent: TEZ-2605
Fix Tez GenericCounter performance bottleneck
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14706229#comment-14706229
]
Gopal V commented on TEZ-2731:
--
[~rajesh.balamohan]: I thought UseBiasedLocking had some issues
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14705660#comment-14705660
]
Gopal V commented on TEZ-2731:
--
[~rajesh.balamohan]: can you review?
Fix Tez GenericCounter
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V reassigned TEZ-2731:
Assignee: Gopal V
Fix Tez GenericCounter performance bottleneck
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2731:
-
Attachment: atomic-long-cntr.png
Fix Tez GenericCounter performance bottleneck
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2731:
-
Attachment: TEZ-2731.1.patch
Fix Tez GenericCounter performance bottleneck
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2731:
-
Attachment: lock-inc.png
mr-reader-next.png
Fix Tez GenericCounter performance bottleneck
Gopal V created TEZ-2673:
Summary: Tez AM logs need to print the container JVM opts with
CONTAINER_STARTED event
Key: TEZ-2673
URL: https://issues.apache.org/jira/browse/TEZ-2673
Project: Apache Tez
[
https://issues.apache.org/jira/browse/TEZ-2673?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14649731#comment-14649731
]
Gopal V commented on TEZ-2673:
--
The stderr has it, but it's impossible to navigate to an stderr
[
https://issues.apache.org/jira/browse/TEZ-2673?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14648716#comment-14648716
]
Gopal V edited comment on TEZ-2673 at 7/31/15 4:02 AM:
---
Possibly, but
[
https://issues.apache.org/jira/browse/TEZ-2673?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14648716#comment-14648716
]
Gopal V commented on TEZ-2673:
--
Possibly, but do you have an easy way to triage
{code}
[
https://issues.apache.org/jira/browse/TEZ-2687?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2687:
-
Attachment: TEZ-2687.1.patch
This is not an explicit dependency, the shutdown order is the reverse order of
[
https://issues.apache.org/jira/browse/TEZ-2687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14654746#comment-14654746
]
Gopal V commented on TEZ-2687:
--
These two seem to be the only ones which block, the re-ordering
Gopal V created TEZ-2663:
Summary: SessionNotRunning exceptions are wrapped in a
ServiceException from a dying AM
Key: TEZ-2663
URL: https://issues.apache.org/jira/browse/TEZ-2663
Project: Apache Tez
[
https://issues.apache.org/jira/browse/TEZ-1997?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14695976#comment-14695976
]
Gopal V commented on TEZ-1997:
--
bq. A parallel spill and collect is possible if I'm not
[
https://issues.apache.org/jira/browse/TEZ-2687?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2687:
-
Assignee: (was: Gopal V)
ATS History shutdown happens before the min-held containers are released
[
https://issues.apache.org/jira/browse/TEZ-2687?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14680881#comment-14680881
]
Gopal V commented on TEZ-2687:
--
Deleting the bad patch attached to the JIRA and leaving the
[
https://issues.apache.org/jira/browse/TEZ-2687?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2687:
-
Attachment: (was: TEZ-2687.1.patch)
ATS History shutdown happens before the min-held containers are released
Gopal V created TEZ-2687:
Summary: ATS History shutdown happens before the min-held
containers are released
Key: TEZ-2687
URL: https://issues.apache.org/jira/browse/TEZ-2687
Project: Apache Tez
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2632:
-
Attachment: TEZ-2632.1.patch
This adds -Paws and -Pazure to the tez build.
As [~hitesh] noted, this will not
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14636349#comment-14636349
]
Gopal V commented on TEZ-2632:
--
Profiles can't be chained and they can't read off a property -
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2632:
-
Summary: A -Paws and -Pazure build profiles for hadoop-{aws,azure}
inclusion (was: A -Ps3 build profile to
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2632:
-
Description:
A number of S3a workloads which work with MRv2 (after HADOOP-10400) is failing
to work with Tez.
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14637697#comment-14637697
]
Gopal V commented on TEZ-2632:
--
Opened a wiki item to track docs -
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14637672#comment-14637672
]
Gopal V commented on TEZ-2632:
--
[~hitesh]: there seems to be no way to enable this on 2
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14637681#comment-14637681
]
Gopal V edited comment on TEZ-2632 at 7/22/15 9:38 PM:
---
bq. s/..., you
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14637681#comment-14637681
]
Gopal V commented on TEZ-2632:
--
bq. s/..., you must explicitly bundle ... for Tez to be
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2632:
-
Component/s: (was: UI)
A -Paws and -Pazure build profiles for hadoop-{aws,azure} inclusion
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2632:
-
Attachment: TEZ-2632.2.patch
Add BUILDING.txt docs
A -Paws and -Pazure build profiles for hadoop-{aws,azure}
[
https://issues.apache.org/jira/browse/TEZ-2602?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14637895#comment-14637895
]
Gopal V commented on TEZ-2602:
--
The patch LGTM - +1.
In a later pass, we can optimize this
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V reassigned TEZ-2632:
Assignee: Gopal V
A -Ps3 build profile to integrate hadoop S3a support
Gopal V created TEZ-2632:
Summary: A -Ps3 build profile to integrate hadoop S3a support
Key: TEZ-2632
URL: https://issues.apache.org/jira/browse/TEZ-2632
Project: Apache Tez
Issue Type: Bug
[
https://issues.apache.org/jira/browse/TEZ-2632?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14635989#comment-14635989
]
Gopal V commented on TEZ-2632:
--
Need to get more of a batteries included, but removable with
[
https://issues.apache.org/jira/browse/TEZ-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14627363#comment-14627363
]
Gopal V commented on TEZ-2606:
--
[~ozawa]: Sure, MR types can also have their own prefix
[
https://issues.apache.org/jira/browse/TEZ-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V resolved TEZ-2606.
--
Resolution: Done
Resolving this issue for now, the prefix key sort is already in 0.7.x and is
available to any
[
https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978640#comment-14978640
]
Gopal V commented on TEZ-2244:
--
[~rajesh.balamohan]: patch LGTM - +1.
I think the config parameter may be
[
https://issues.apache.org/jira/browse/TEZ-2244?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14978791#comment-14978791
]
Gopal V commented on TEZ-2244:
--
No, it doesn't.
The existing behaviour is as follows - the system will not
[
https://issues.apache.org/jira/browse/TEZ-2879?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14992889#comment-14992889
]
Gopal V commented on TEZ-2879:
--
[~sseth]: the API LGTM - +1.
This mechanism for location pinning can also
Gopal V created TEZ-2891:
Summary: Hardcoded timeouts: Shortest Tez Query possible is 500ms
Key: TEZ-2891
URL: https://issues.apache.org/jira/browse/TEZ-2891
Project: Apache Tez
Issue Type: Bug
[
https://issues.apache.org/jira/browse/TEZ-2891?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V resolved TEZ-2891.
--
Resolution: Invalid
> Hardcoded timeouts: Shortest Tez Query possible is 500ms
>
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14960176#comment-14960176
]
Gopal V commented on TEZ-2849:
--
[~karams]: Can you test this patch with terasort on JDK7 and JDK8?
[~ozawa]:
[
https://issues.apache.org/jira/browse/TEZ-2850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14961474#comment-14961474
]
Gopal V commented on TEZ-2850:
--
[~jeagles]: good call on the in == null check, that's a valid assumption for
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14946109#comment-14946109
]
Gopal V commented on TEZ-2849:
--
[~ozawa]: the patch works for me on JDK8 - can you tell me what's the actual
Gopal V created TEZ-2881:
Summary: Fix the InputBuffer bottleneck during the in-memory spill
merge
Key: TEZ-2881
URL: https://issues.apache.org/jira/browse/TEZ-2881
Project: Apache Tez
Issue Type:
[
https://issues.apache.org/jira/browse/TEZ-2881?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V resolved TEZ-2881.
--
Resolution: Not A Problem
> Fix the InputBuffer bottleneck during the in-memory spill merge
>
Gopal V created TEZ-2889:
Summary: ATSHistoryACLPolicyManager makes DAG Submission wait on
ATS requests
Key: TEZ-2889
URL: https://issues.apache.org/jira/browse/TEZ-2889
Project: Apache Tez
Issue
[
https://issues.apache.org/jira/browse/TEZ-2889?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2889:
-
Labels: ATS Performance (was: )
> ATSHistoryACLPolicyManager makes DAG Submission wait on ATS requests
>
[
https://issues.apache.org/jira/browse/TEZ-2889?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2889:
-
Assignee: Hitesh Shah
> ATSHistoryACLPolicyManager makes DAG Submission wait on ATS requests
>
[
https://issues.apache.org/jira/browse/TEZ-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14619142#comment-14619142
]
Gopal V commented on TEZ-2606:
--
[~ozawa]: I think I implemented this sometime in late 2013 -
[
https://issues.apache.org/jira/browse/TEZ-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14621065#comment-14621065
]
Gopal V commented on TEZ-2606:
--
bq. My idea is kvmeta's including new prefix of key's binary.
[
https://issues.apache.org/jira/browse/TEZ-2606?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14621075#comment-14621075
]
Gopal V edited comment on TEZ-2606 at 7/9/15 6:58 PM:
--
bq. And even
[
https://issues.apache.org/jira/browse/TEZ-2610?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14621636#comment-14621636
]
Gopal V commented on TEZ-2610:
--
The patch seems to work just fine, between releases too -
[
https://issues.apache.org/jira/browse/TEZ-2610?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V reassigned TEZ-2610:
Assignee: Gopal V
Swimlane for DAGs that use containers from previous DAGs
[
https://issues.apache.org/jira/browse/TEZ-2610?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2610:
-
Attachment: TEZ-2610.patch
[~yeeza]: preliminary patch - this should work, still refining the view
Swimlane for
[
https://issues.apache.org/jira/browse/TEZ-2731?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2731:
-
Fix Version/s: 0.8.0
Fix Tez GenericCounter performance bottleneck
-
[
https://issues.apache.org/jira/browse/TEZ-2741?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V moved HIVE-11396 to TEZ-2741:
-
Target Version/s: 0.8.0
Key: TEZ-2741 (was: HIVE-11396)
[
https://issues.apache.org/jira/browse/TEZ-2741?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2741:
-
Attachment: TEZ-2741.1.patch
Satisfy old MRv1 api's contract, by calling createKey() createValue() when
switching
[
https://issues.apache.org/jira/browse/TEZ-1491?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-1491:
-
Attachment: TEZ-1491.1.patch
Use a single writer override for the Progress.set().
This is confined to two parts of
[
https://issues.apache.org/jira/browse/TEZ-2834?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14790783#comment-14790783
]
Gopal V commented on TEZ-2834:
--
[~bikassaha]: YARN-4149? That was fixed last night, it's not deployed on the
[
https://issues.apache.org/jira/browse/TEZ-2850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14935959#comment-14935959
]
Gopal V commented on TEZ-2850:
--
I've been trying to understand why we even have a reference to IFileInputStream
[
https://issues.apache.org/jira/browse/TEZ-2834?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14876693#comment-14876693
]
Gopal V commented on TEZ-2834:
--
Rolling the patch into the weekend tests - do you want to flip any logs to INFO
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2849:
-
Description:
The CPU performance overhead of the sort operation is tied to the JIT handling
of the aliased byte
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V reassigned TEZ-2849:
Assignee: Gopal V
> Implement Specific Workaround for JDK-8026049 & JDK-8073093
>
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2849:
-
Affects Version/s: 0.8.1
> Implement Specific Workaround for JDK-8026049 & JDK-8073093
>
[
https://issues.apache.org/jira/browse/TEZ-2850?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=14903821#comment-14903821
]
Gopal V commented on TEZ-2850:
--
Good catch [~saikatr], that's 4kb of space overhead for 100 bytes of data.
The
[
https://issues.apache.org/jira/browse/TEZ-2849?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2849:
-
Attachment: TEZ-2849.1.patch
Replace the slow parts of the ByteBuffer operations with a disjoint array copy
[
https://issues.apache.org/jira/browse/TEZ-3014?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15064751#comment-15064751
]
Gopal V commented on TEZ-3014:
--
[~bikassaha]: upload the hprof OOM heapdump?
> OOM during Shuffle in JDK 8
>
[
https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15029792#comment-15029792
]
Gopal V commented on TEZ-2964:
--
Had to go to 100k events to fix this at max concurrency
> Tez AMScheduler -
[
https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15029542#comment-15029542
]
Gopal V commented on TEZ-2964:
--
This looks like an AM recovery related issue, because I see 2 AM attempts for
[
https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Gopal V updated TEZ-2964:
-
Attachment: TEZ-2964.tgz
> Tez AMScheduler - assign to unallocated container
>
Gopal V created TEZ-2964:
Summary: Tez AMScheduler - assign to unallocated container
Key: TEZ-2964
URL: https://issues.apache.org/jira/browse/TEZ-2964
Project: Apache Tez
Issue Type: Bug
Affects
[
https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15029727#comment-15029727
]
Gopal V commented on TEZ-2964:
--
Disabling DAG recovery fixes this error.
Ok, so I counted about 11,000
[
https://issues.apache.org/jira/browse/TEZ-2964?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15029743#comment-15029743
]
Gopal V commented on TEZ-2964:
--
Yes, that fixes the issue, thanks [~rajesh.balamohan]
> Tez AMScheduler -
[
https://issues.apache.org/jira/browse/TEZ-2950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15012291#comment-15012291
]
Gopal V commented on TEZ-2950:
--
[~rohini]: this is already implemented for UnorderedPartitioned, right?
set
[
https://issues.apache.org/jira/browse/TEZ-2950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15012429#comment-15012429
]
Gopal V commented on TEZ-2950:
--
bq, enable final merge in output = false doesn't necessarily solve this. That
[
https://issues.apache.org/jira/browse/TEZ-2950?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=15012371#comment-15012371
]
Gopal V commented on TEZ-2950:
--
That is a per-edge configuration.
> Poor performance of
201 - 300 of 521 matches
Mail list logo