[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-05-04 Thread Siddharth Seth (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14527068#comment-14527068
 ] 

Siddharth Seth commented on TEZ-1897:
-

Looks like this went in with concurrentDispatchers enabled. Can you please undo 
that bit.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Fix For: 0.7.0

 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch, TEZ-1897.7.patch, 
 TEZ-1897.8.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-05-04 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14527081#comment-14527081
 ] 

Bikas Saha commented on TEZ-1897:
-

Thanks for catching it. My bad. Fixed.
commit 5218f481dba2a26c3aa5dd8f69285ab9da419dd1
Author: Bikas Saha bi...@apache.org
Date:   Mon May 4 12:05:39 2015 -0700

TEZ-1897 addendum to turn off by default . Create a concurrent version of 
AsyncDispatcher (bikas)



 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Fix For: 0.7.0

 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch, TEZ-1897.7.patch, 
 TEZ-1897.8.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-05-01 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14523499#comment-14523499
 ] 

Bikas Saha commented on TEZ-1897:
-

bq. AsyncDispatcherConcurrent(String name, int numThreads) { super(name) 
instead of super(dispatcher)
bq. final LinkedBlockingQueueEvent queue;; - Double ;
bq. In AsyncDispatcher - the error checking code for previously registered 
dispatchers
bq. aitForDrained.wait(1000);, LO
Fixed

bq. At the same place - do the threads need to 
Cannot interrupt executorservice. After waiting shutdownNow is called which 
should take care of this. In any case, drained code patch is not used and we 
should probably remove it separately.

bq. serviceStop / serviceStart don't need to invoke super. 
Thats the way other CompositeServices are written and seems to be the correct 
thing to do since this method overrides super.serviceStart().

bq. here's a lot of code duplication between AsyncDispatcher and 
AsyncDispatcherConcu
I chose to do so, so that the concurrent version can replace the existing 
version in a future release. The concurrent version with thread size set to 1 
is the same as the existing version. Manual thread creation is replaced by 
using a single thread in the threadpool. So there is no need to maintain 2 
classes.

thanks for the reviews. Unless there are further comments, will commit by EOD.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-05-01 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14523641#comment-14523641
 ] 

Bikas Saha commented on TEZ-1897:
-

From what I see - AbstractService#start()-serviceStart()-end up calling 
AsyncDispatcher-serviceStart() since it overrides 
CompositeService#serviceStart(). So the derived class must call its parents 
serviceStart(). MRAppMaster is also written that way.
Thanks! We can remove this draining code and other unused code when we remove 
AsyncDispatcher.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch, TEZ-1897.7.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-05-01 Thread TezQA (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14523659#comment-14523659
 ] 

TezQA commented on TEZ-1897:


{color:red}-1 overall{color}.  Here are the results of testing the latest 
attachment
  http://issues.apache.org/jira/secure/attachment/12729796/TEZ-1897.7.patch
  against master revision c96eed3.

{color:green}+1 @author{color}.  The patch does not contain any @author 
tags.

{color:green}+1 tests included{color}.  The patch appears to include 7 new 
or modified test files.

{color:green}+1 javac{color}.  The applied patch does not increase the 
total number of javac compiler warnings.

{color:green}+1 javadoc{color}.  There were no new javadoc warning messages.

{color:green}+1 findbugs{color}.  The patch does not introduce any new 
Findbugs (version 2.0.3) warnings.

{color:red}-1 release audit{color}.  The applied patch generated 1 
release audit warnings.

{color:green}+1 core tests{color}.  The patch passed unit tests in .

Test results: https://builds.apache.org/job/PreCommit-TEZ-Build/602//testReport/
Release audit warnings: 
https://builds.apache.org/job/PreCommit-TEZ-Build/602//artifact/patchprocess/patchReleaseAuditProblems.txt
Console output: https://builds.apache.org/job/PreCommit-TEZ-Build/602//console

This message is automatically generated.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch, TEZ-1897.7.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-30 Thread Siddharth Seth (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14522300#comment-14522300
 ] 

Siddharth Seth commented on TEZ-1897:
-

Rest looks good to me.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-30 Thread Siddharth Seth (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14522298#comment-14522298
 ] 

Siddharth Seth commented on TEZ-1897:
-

Mostly minor comments.
- AsyncDispatcherConcurrent(String name, int numThreads) { super(name) 
instead of super(dispatcher)
- final LinkedBlockingQueueEvent queue;; - Double ;
- serviceStop / serviceStart don't need to invoke super. These will be invoked 
via the stop() / start() methods automatically.
- In AsyncDispatcher - the error checking code for previously registered 
dispatchers (/* check to see if we have a listener registered */) can be 
common across all register methods. registerAndCreateDispatcher(without 
#threads) is missing a check on ConcurrentDispatchers
- waitForDrained.wait(1000);, LOG.info(Waiting for AsyncDispatcher to 
drain.); - log line before the wait ?
- At the same place - do the threads need to be interrupted ? Otherwise they'll 
always wait for the 1000ms if the queue is already empty.
- There's changes in TaskImpl, Task, Vertex - which are unrelated to this. 
That's likely adding to perf gains. There's a separate jira for this - will see 
if the scope of that is no longer valid after theses changes.

Major bit: 
There's a lot of code duplication between AsyncDispatcher and 
AsyncDispatcherConcurrent - GenericEventHandler, MultiAttemptListener and a 
bunch more.The register functions seem a little unnecessary since noone 
registers directly with this. Not sure if you want to fix this here or in a 
follow up.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14518254#comment-14518254
 ] 

Bikas Saha commented on TEZ-1897:
-

bq. Rename registerAndCreateDispatcher to registerAndCreateConcurrentDispatcher 
?
bq. Rename registerWithExistingDispatcher to 
registerWithExistingConcurrentDispatcher ?
The types in the arguments should be enough to differentiate and also provide 
compile time check. Adding it to the names would be redundant. Also, once this 
code is exercised and stabilized then we may use this as the only dispatcher.

bq. Javadoc for AsyncDispatcherConcurrent is the same as AsyncDispatcher
Fixed

bq. Precondition numThreads should be at least 2 ? I think 1 is the same as 
AysnDispatcher
This should be able to replace the legacy AsyncDispatcher and run all events on 
a single thread in the threadpool. So its allowed.

bq. TaskAttemptImpl.java Some places still use getTask().getVertex(), can 
change to getVertex directly.
Fixed


 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread TezQA (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14518493#comment-14518493
 ] 

TezQA commented on TEZ-1897:


{color:green}+1 overall{color}.  Here are the results of testing the latest 
attachment
  http://issues.apache.org/jira/secure/attachment/12728962/TEZ-1897.6.patch
  against master revision 5b2f011.

{color:green}+1 @author{color}.  The patch does not contain any @author 
tags.

{color:green}+1 tests included{color}.  The patch appears to include 6 new 
or modified test files.

{color:green}+1 javac{color}.  The applied patch does not increase the 
total number of javac compiler warnings.

{color:green}+1 javadoc{color}.  There were no new javadoc warning messages.

{color:green}+1 findbugs{color}.  The patch does not introduce any new 
Findbugs (version 2.0.3) warnings.

{color:green}+1 release audit{color}.  The applied patch does not increase 
the total number of release audit warnings.

{color:green}+1 core tests{color}.  The patch passed unit tests in .

Test results: https://builds.apache.org/job/PreCommit-TEZ-Build/567//testReport/
Console output: https://builds.apache.org/job/PreCommit-TEZ-Build/567//console

This message is automatically generated.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.6.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Siddharth Seth (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14517978#comment-14517978
 ] 

Siddharth Seth commented on TEZ-1897:
-

Quick glance. This should be disabled by default.
Will look in detail a little later.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14517996#comment-14517996
 ] 

Bikas Saha commented on TEZ-1897:
-

the patch enabled it by default to get the jenkins run with it. It will be off 
by default.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Jeff Zhang (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14516643#comment-14516643
 ] 

Jeff Zhang commented on TEZ-1897:
-

Comments:
* Rename registerAndCreateDispatcher to registerAndCreateConcurrentDispatcher ?
  Rename registerWithExistingDispatcher to 
registerWithExistingConcurrentDispatcher ?

* Javadoc for AsyncDispatcherConcurrent is the same as AsyncDispatcher

* No concurentDispatcher checking in registerAndCreateDispatcher(Class? 
extends Enum eventType,
  EventHandler handler, String dispatcherName)

*  Precondition numThreads should be at least 2 ? I think 1 is the same as 
AysnDispatcher
{code}
 public AsyncDispatcherConcurrent registerAndCreateDispatcher(Class? extends 
Enum eventType,
  EventHandler handler, String dispatcherName, int numThreads) 
{code}

* TaskAttemptImpl.java
Some places still use getTask().getVertex(), can change to getVertex directly.



 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14516578#comment-14516578
 ] 

Bikas Saha commented on TEZ-1897:
-

Sorry about that. Deleted the wrong one.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14516528#comment-14516528
 ] 

Bikas Saha commented on TEZ-1897:
-

Folks, please review. I would like to get this into 0.7. Thanks!

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-28 Thread Jeff Zhang (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14516530#comment-14516530
 ] 

Jeff Zhang commented on TEZ-1897:
-

Which one is the latest patch ? Saw 2 TEZ-1897.5.patch and both 5 days ago.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-23 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14510002#comment-14510002
 ] 

Bikas Saha commented on TEZ-1897:
-

[~zjffdu] [~hitesh] Please continue your reviews with the latest patch.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-22 Thread TezQA (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14507588#comment-14507588
 ] 

TezQA commented on TEZ-1897:


{color:red}-1 overall{color}.  Here are the results of testing the latest 
attachment
  http://issues.apache.org/jira/secure/attachment/12727292/TEZ-1897.4.patch
  against master revision 9faa54c.

{color:green}+1 @author{color}.  The patch does not contain any @author 
tags.

{color:green}+1 tests included{color}.  The patch appears to include 6 new 
or modified test files.

{color:green}+1 javac{color}.  The applied patch does not increase the 
total number of javac compiler warnings.

{color:green}+1 javadoc{color}.  There were no new javadoc warning messages.

{color:red}-1 findbugs{color}.  The patch appears to introduce 2 new 
Findbugs (version 2.0.3) warnings.

{color:green}+1 release audit{color}.  The applied patch does not increase 
the total number of release audit warnings.

{color:red}-1 core tests{color}.  The patch failed these unit tests in :
   org.apache.tez.dag.app.TestMockDAGAppMaster

Test results: https://builds.apache.org/job/PreCommit-TEZ-Build/514//testReport/
Findbugs warnings: 
https://builds.apache.org/job/PreCommit-TEZ-Build/514//artifact/patchprocess/newPatchFindbugsWarningstez-common.html
Console output: https://builds.apache.org/job/PreCommit-TEZ-Build/514//console

This message is automatically generated.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-22 Thread Bikas Saha (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14508247#comment-14508247
 ] 

Bikas Saha commented on TEZ-1897:
-

The last patch builds on the previous patch to actually use the concurrent 
dispatcher to run Task and TaskAttempt events concurrently. There is a 
configuration to turn this on or off and when it is turned off the code runs 
exactly the same path as it does today. So this change is very safe.

In order to keep things sane, events for a given Task and its attempts are 
serialized on the same thread. This is done by using a serializing hash 
determined from the TezTaskID. Different tasks run on different threads. That 
takes care of a lot of locking issues. Next, Vertex has reference to DAG, Task 
has reference to Vertex and Attempt has reference to Task and Vertex. This 
helps remove unnecessary locking issues and delays that occur when they are 
accessed from the AppContext to get dag/vertex/task etc. and then look-up into 
their internal maps. This change would be beneficial in general by reducing 
lock contention compared to today.

Added a simulation test that runs 5 tasks at 1000 concurrency which runs up 
to 30% faster with the change than without.

The patch has the config turned on for patch test execution. This will be off 
by default and is marked private so only advanced users can try this for large 
clusters where we can get 10-20K running tasks concurrently.

Please review.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)


[jira] [Commented] (TEZ-1897) Create a concurrent version of AsyncDispatcher

2015-04-22 Thread TezQA (JIRA)

[ 
https://issues.apache.org/jira/browse/TEZ-1897?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14508132#comment-14508132
 ] 

TezQA commented on TEZ-1897:


{color:green}+1 overall{color}.  Here are the results of testing the latest 
attachment
  http://issues.apache.org/jira/secure/attachment/12727349/TEZ-1897.5.patch
  against master revision 9faa54c.

{color:green}+1 @author{color}.  The patch does not contain any @author 
tags.

{color:green}+1 tests included{color}.  The patch appears to include 6 new 
or modified test files.

{color:green}+1 javac{color}.  The applied patch does not increase the 
total number of javac compiler warnings.

{color:green}+1 javadoc{color}.  There were no new javadoc warning messages.

{color:green}+1 findbugs{color}.  The patch does not introduce any new 
Findbugs (version 2.0.3) warnings.

{color:green}+1 release audit{color}.  The applied patch does not increase 
the total number of release audit warnings.

{color:green}+1 core tests{color}.  The patch passed unit tests in .

Test results: https://builds.apache.org/job/PreCommit-TEZ-Build/516//testReport/
Console output: https://builds.apache.org/job/PreCommit-TEZ-Build/516//console

This message is automatically generated.

 Create a concurrent version of AsyncDispatcher
 --

 Key: TEZ-1897
 URL: https://issues.apache.org/jira/browse/TEZ-1897
 Project: Apache Tez
  Issue Type: Task
Reporter: Bikas Saha
Assignee: Bikas Saha
 Attachments: TEZ-1897.1.patch, TEZ-1897.2.patch, TEZ-1897.3.patch, 
 TEZ-1897.4.patch, TEZ-1897.5.patch, TEZ-1897.5.patch


 Currently, it processes events on a single thread. For events that can be 
 executed in parallel, e.g. vertex manager events, allowing higher concurrency 
 may be beneficial.



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)