[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to Netty4 and remove Netty3 dependency from hive where it's possible
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=586405=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-586405 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 21/Apr/21 08:13 Start Date: 21/Apr/21 08:13 Worklog Time Spent: 10m Work Description: abstractdog commented on pull request #1778: URL: https://github.com/apache/hive/pull/1778#issuecomment-823871712 merged to master, thanks @pgaref for the review! -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 586405) Time Spent: 2h 50m (was: 2h 40m) > LLAP ShuffleHandler: upgrade to Netty4 and remove Netty3 dependency from hive > where it's possible > - > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Labels: pull-request-available > Fix For: 4.0.0 > > Attachments: dep.log > > Time Spent: 2h 50m > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is a possible performance improvement compared to Netty3. However, the > refactor is not trivial, TEZ-4157 covers that more or less (the code bases > are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ > some other notes: Netty3 is EOL since 2016: > https://netty.io/news/2016/06/29/3-10-6-Final.html -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to Netty4 and remove Netty3 dependency from hive where it's possible
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=586404=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-586404 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 21/Apr/21 08:13 Start Date: 21/Apr/21 08:13 Worklog Time Spent: 10m Work Description: abstractdog merged pull request #1778: URL: https://github.com/apache/hive/pull/1778 -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 586404) Time Spent: 2h 40m (was: 2.5h) > LLAP ShuffleHandler: upgrade to Netty4 and remove Netty3 dependency from hive > where it's possible > - > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Labels: pull-request-available > Attachments: dep.log > > Time Spent: 2h 40m > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is a possible performance improvement compared to Netty3. However, the > refactor is not trivial, TEZ-4157 covers that more or less (the code bases > are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ > some other notes: Netty3 is EOL since 2016: > https://netty.io/news/2016/06/29/3-10-6-Final.html -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581698=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581698
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 11:06
Start Date: 13/Apr/21 11:06
Worklog Time Spent: 10m
Work Description: pgaref commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612347286
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
Thanks Laszlo! sounds like a plan!
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581698)
Time Spent: 2.5h (was: 2h 20m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 2.5h
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581694=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581694
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 11:00
Start Date: 13/Apr/21 11:00
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612343543
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
okay, in this case I'll have to include some unit tests here (which are
part of tez codebase already) + create a simple repro to share with netty
community
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581694)
Time Spent: 2h 20m (was: 2h 10m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 2h 20m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581692=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581692
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 11:00
Start Date: 13/Apr/21 11:00
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612343543
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
okay, in this case I'll have to include some unit tests here (which
might be part of tez code already) + create a simple repro to share with netty
community
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581692)
Time Spent: 2h 10m (was: 2h)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 2h 10m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581684=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581684
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:44
Start Date: 13/Apr/21 10:44
Worklog Time Spent: 10m
Work Description: pgaref commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612334078
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
Got it, this is helpful but lets make sure this is expected from nettys'
side of things before committing -- this would be helpful for the Tez change as
well :)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581684)
Time Spent: 2h (was: 1h 50m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 2h
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581683=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581683
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:38
Start Date: 13/Apr/21 10:38
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612330581
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
good catch :) this is an epic workaround for a problem that I haven't
been able to figure out 100%, here are some details:
https://issues.apache.org/jira/browse/TEZ-4157?focusedCommentId=17100835=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-17100835
(btw: with netty3, we didn't need this)
are you fine with a comment explaining this?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581683)
Time Spent: 1h 50m (was: 1h 40m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1h 50m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581682=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581682
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:38
Start Date: 13/Apr/21 10:38
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612330581
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -797,16 +803,17 @@ public void messageReceived(ChannelHandlerContext ctx,
MessageEvent evt)
Map mapOutputInfoMap =
new HashMap();
- Channel ch = evt.getChannel();
-
+ Channel ch = ctx.channel();
// In case of KeepAlive, ensure that timeout handler does not close
connection until entire
// response is written (i.e, response headers + mapOutput).
- ChannelPipeline pipeline = ch.getPipeline();
+ ChannelPipeline pipeline = ch.pipeline();
TimeoutHandler timeoutHandler =
(TimeoutHandler)pipeline.get(TIMEOUT_HANDLER);
timeoutHandler.setEnabledTimeout(false);
String user = userRsrc.get(jobId);
-
+ if (keepAliveParam || connectionKeepAliveEnabled){
Review comment:
good catch :) this is an epic workaround that I haven't been able to
figure out, here are some details:
https://issues.apache.org/jira/browse/TEZ-4157?focusedCommentId=17100835=com.atlassian.jira.plugin.system.issuetabpanels%3Acomment-tabpanel#comment-17100835
(btw: with netty3, we didn't need this)
are you fine with a comment explaining this?
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581682)
Time Spent: 1h 40m (was: 1.5h)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1h 40m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581680=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581680
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:36
Start Date: 13/Apr/21 10:36
Worklog Time Spent: 10m
Work Description: pgaref commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612329298
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/FadvisedFileRegion.java
##
@@ -71,15 +72,39 @@ public long transferTo(WritableByteChannel target, long
position)
throws IOException {
if (manageOsCache && readaheadPool != null) {
readaheadRequest = readaheadPool.readaheadStream(identifier, fd,
- getPosition() + position, readaheadLength,
- getPosition() + getCount(), readaheadRequest);
+ position() + position, readaheadLength,
+ position() + count(), readaheadRequest);
}
-
+long written = 0;
Review comment:
Got it
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581680)
Time Spent: 1.5h (was: 1h 20m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1.5h
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581679=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581679
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:34
Start Date: 13/Apr/21 10:34
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612328280
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/FadvisedFileRegion.java
##
@@ -71,15 +72,39 @@ public long transferTo(WritableByteChannel target, long
position)
throws IOException {
if (manageOsCache && readaheadPool != null) {
readaheadRequest = readaheadPool.readaheadStream(identifier, fd,
- getPosition() + position, readaheadLength,
- getPosition() + getCount(), readaheadRequest);
+ position() + position, readaheadLength,
+ position() + count(), readaheadRequest);
}
-
+long written = 0;
Review comment:
looks better, but I don't think it's correct: in case of an exception
during the transfer, we should not have set transferred=true
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581679)
Time Spent: 1h 20m (was: 1h 10m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1h 20m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581677=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581677
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:31
Start Date: 13/Apr/21 10:31
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612326445
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/FadvisedFileRegion.java
##
@@ -124,39 +149,33 @@ long customShuffleTransfer(WritableByteChannel target,
long position)
position += trans;
trans = 0;
}
-
+
//write data to the target
while(byteBuffer.hasRemaining()) {
target.write(byteBuffer);
}
byteBuffer.clear();
}
-
+
return actualCount - trans;
}
-
- @Override
- public void releaseExternalResources() {
-if (readaheadRequest != null) {
- readaheadRequest.cancel();
-}
-super.releaseExternalResources();
- }
-
/**
* Call when the transfer completes successfully so we can advise the OS that
* we don't need the region to be cached anymore.
*/
public void transferSuccessful() {
-if (manageOsCache && getCount() > 0) {
+if (manageOsCache && count() > 0) {
try {
if (canEvictAfterTransfer) {
- LOG.debug("shuffleBufferSize: {}, path: {}", shuffleBufferSize,
identifier);
-
NativeIO.POSIX.getCacheManipulator().posixFadviseIfPossible(identifier,
- fd, getPosition(), getCount(),
- NativeIO.POSIX.POSIX_FADV_DONTNEED);
+ if (fd.valid()) {
Review comment:
hm, thought this over again, fd.valid() change was needed while I
haven't been handling deallocate() stuff properly, but now, at this point fd
should be valid...initially I left this check here because I thought that an
invalid fd is not a problem (which is true, we won't advise to OS cache, and
that's it), but as we already have try/catch, we don't need this this check
(we'll have the exception in the logs anyway)
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581677)
Time Spent: 1h 10m (was: 1h)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1h 10m
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581674=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581674
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:27
Start Date: 13/Apr/21 10:27
Worklog Time Spent: 10m
Work Description: abstractdog commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612323669
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -339,27 +350,60 @@ private ShuffleHandler(Configuration conf) {
public void start() throws Exception {
-ServerBootstrap bootstrap = new ServerBootstrap(selector);
-// Timer is shared across entire factory and must be released separately
-timer = new HashedWheelTimer();
-try {
- pipelineFact = new HttpPipelineFactory(conf, timer);
-} catch (Exception ex) {
- throw new RuntimeException(ex);
-}
-bootstrap.setPipelineFactory(pipelineFact);
-bootstrap.setOption("backlog", NetUtil.SOMAXCONN);
+ServerBootstrap bootstrap = new ServerBootstrap()
+.channel(NioServerSocketChannel.class)
+.group(bossGroup, workerGroup)
+.localAddress(port)
+.option(ChannelOption.SO_BACKLOG, NetUtil.SOMAXCONN)
+.childOption(ChannelOption.SO_KEEPALIVE, true);
+initPipeline(bootstrap, conf);
+
port = conf.getInt(SHUFFLE_PORT_CONFIG_KEY, DEFAULT_SHUFFLE_PORT);
-Channel ch = bootstrap.bind(new InetSocketAddress(port));
+Channel ch = bootstrap.bind().sync().channel();
accepted.add(ch);
-port = ((InetSocketAddress)ch.getLocalAddress()).getPort();
+port = ((InetSocketAddress)ch.localAddress()).getPort();
conf.set(SHUFFLE_PORT_CONFIG_KEY, Integer.toString(port));
-pipelineFact.SHUFFLE.setPort(port);
+SHUFFLE.setPort(port);
if (dirWatcher != null) {
dirWatcher.start();
}
-LOG.info("LlapShuffleHandler" + " listening on port " + port + "
(SOMAXCONN: " + bootstrap.getOption("backlog")
- + ")");
+LOG.info("LlapShuffleHandler listening on port {} (SOMAXCONN: {})", port,
NetUtil.SOMAXCONN);
+ }
+
+ private void initPipeline(ServerBootstrap bootstrap, Configuration conf)
throws Exception {
+SHUFFLE = getShuffle(conf);
+// TODO Setup SSL Shuffle
Review comment:
I think we don't support SSL shuffle for LLAP at the moment (+ the
comment is quite old), e.g. Cloudera's data warehouse, ssl on shuffle is
handled transparently by the environment
I haven't touched this part in this patch, and not even sure what's the plan
:) that's why I simply kept this as is
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
Issue Time Tracking
---
Worklog Id: (was: 581674)
Time Spent: 1h (was: 50m)
> LLAP ShuffleHandler: upgrade to netty4
> --
>
> Key: HIVE-24524
> URL: https://issues.apache.org/jira/browse/HIVE-24524
> Project: Hive
> Issue Type: Improvement
>Reporter: László Bodor
>Assignee: László Bodor
>Priority: Major
> Labels: pull-request-available
> Time Spent: 1h
> Remaining Estimate: 0h
>
> Tez already has a WIP patch for upgrading its shuffle handler to netty4.
> Netty4 is told to be a possible performance improvement compared to Netty3.
> However, the refactor is not trivial, TEZ-4157 covers that more or less (the
> code bases are very similar).
> Background:
> netty4 migration guideline:
> https://netty.io/wiki/new-and-noteworthy-in-4.0.html
> articles of possible performance improvement:
> https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html
> https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/
> some other notes: Netty3 is EOL since 2016:
> https://netty.io/news/2016/06/29/3-10-6-Final.html
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[
https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=581667=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-581667
]
ASF GitHub Bot logged work on HIVE-24524:
-
Author: ASF GitHub Bot
Created on: 13/Apr/21 10:11
Start Date: 13/Apr/21 10:11
Worklog Time Spent: 10m
Work Description: pgaref commented on a change in pull request #1778:
URL: https://github.com/apache/hive/pull/1778#discussion_r612289762
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/FadvisedFileRegion.java
##
@@ -71,15 +72,39 @@ public long transferTo(WritableByteChannel target, long
position)
throws IOException {
if (manageOsCache && readaheadPool != null) {
readaheadRequest = readaheadPool.readaheadStream(identifier, fd,
- getPosition() + position, readaheadLength,
- getPosition() + getCount(), readaheadRequest);
+ position() + position, readaheadLength,
+ position() + count(), readaheadRequest);
}
-
+long written = 0;
Review comment:
Shall we simplify this to:
```
transferred = true;
if (this.shuffleTransferToAllowed) {
return super.transferTo(target, position);
}
return customShuffleTransfer(target, position);
```
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -339,27 +350,60 @@ private ShuffleHandler(Configuration conf) {
public void start() throws Exception {
-ServerBootstrap bootstrap = new ServerBootstrap(selector);
-// Timer is shared across entire factory and must be released separately
-timer = new HashedWheelTimer();
-try {
- pipelineFact = new HttpPipelineFactory(conf, timer);
-} catch (Exception ex) {
- throw new RuntimeException(ex);
-}
-bootstrap.setPipelineFactory(pipelineFact);
-bootstrap.setOption("backlog", NetUtil.SOMAXCONN);
+ServerBootstrap bootstrap = new ServerBootstrap()
+.channel(NioServerSocketChannel.class)
+.group(bossGroup, workerGroup)
+.localAddress(port)
+.option(ChannelOption.SO_BACKLOG, NetUtil.SOMAXCONN)
+.childOption(ChannelOption.SO_KEEPALIVE, true);
+initPipeline(bootstrap, conf);
+
port = conf.getInt(SHUFFLE_PORT_CONFIG_KEY, DEFAULT_SHUFFLE_PORT);
-Channel ch = bootstrap.bind(new InetSocketAddress(port));
+Channel ch = bootstrap.bind().sync().channel();
accepted.add(ch);
-port = ((InetSocketAddress)ch.getLocalAddress()).getPort();
+port = ((InetSocketAddress)ch.localAddress()).getPort();
conf.set(SHUFFLE_PORT_CONFIG_KEY, Integer.toString(port));
-pipelineFact.SHUFFLE.setPort(port);
+SHUFFLE.setPort(port);
if (dirWatcher != null) {
dirWatcher.start();
}
-LOG.info("LlapShuffleHandler" + " listening on port " + port + "
(SOMAXCONN: " + bootstrap.getOption("backlog")
- + ")");
+LOG.info("LlapShuffleHandler listening on port {} (SOMAXCONN: {})", port,
NetUtil.SOMAXCONN);
+ }
+
+ private void initPipeline(ServerBootstrap bootstrap, Configuration conf)
throws Exception {
+SHUFFLE = getShuffle(conf);
+// TODO Setup SSL Shuffle
Review comment:
I know this is copy pasted from below but do we have a ticket for this?
Is it still needed?
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/ShuffleHandler.java
##
@@ -1031,25 +1038,14 @@ protected ChannelFuture
sendMapOutput(ChannelHandlerContext ctx, Channel ch,
info.getStartOffset(), info.getPartLength(), manageOsCache,
readaheadLength,
readaheadPool, spillfile.getAbsolutePath(),
shuffleBufferSize, shuffleTransferToAllowed,
canEvictAfterTransfer);
-writeFuture = ch.write(partition);
-writeFuture.addListener(new ChannelFutureListener() {
-// TODO error handling; distinguish IO/connection failures,
-// attribute to appropriate spill output
- @Override
- public void operationComplete(ChannelFuture future) {
-if (future.isSuccess()) {
- partition.transferSuccessful();
-}
-partition.releaseExternalResources();
- }
-});
+writeFuture = ch.writeAndFlush(partition);
Review comment:
This looks much cleaner with deallocate() call replacing completion
Listeners
##
File path:
llap-server/src/java/org/apache/hadoop/hive/llap/shufflehandler/FadvisedFileRegion.java
##
@@ -124,39 +149,33 @@ long customShuffleTransfer(WritableByteChannel target,
long position)
position += trans;
trans = 0;
}
-
+
//write data to the target
while(byteBuffer.hasRemaining()) {
target.write(byteBuffer);
}
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=579929=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-579929 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 09/Apr/21 11:53 Start Date: 09/Apr/21 11:53 Worklog Time Spent: 10m Work Description: abstractdog opened a new pull request #1778: URL: https://github.com/apache/hive/pull/1778 ### What changes were proposed in this pull request? ### Why are the changes needed? ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 579929) Time Spent: 40m (was: 0.5h) > LLAP ShuffleHandler: upgrade to netty4 > -- > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Labels: pull-request-available > Time Spent: 40m > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is told to be a possible performance improvement compared to Netty3. > However, the refactor is not trivial, TEZ-4157 covers that more or less (the > code bases are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ > some other notes: Netty3 is EOL since 2016: > https://netty.io/news/2016/06/29/3-10-6-Final.html -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=566659=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-566659 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 16/Mar/21 00:50 Start Date: 16/Mar/21 00:50 Worklog Time Spent: 10m Work Description: github-actions[bot] closed pull request #1778: URL: https://github.com/apache/hive/pull/1778 This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 566659) Time Spent: 0.5h (was: 20m) > LLAP ShuffleHandler: upgrade to netty4 > -- > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Labels: pull-request-available > Time Spent: 0.5h > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is told to be a possible performance improvement compared to Netty3. > However, the refactor is not trivial, TEZ-4157 covers that more or less (the > code bases are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ > some other notes: Netty3 is EOL since 2016: > https://netty.io/news/2016/06/29/3-10-6-Final.html -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=562083=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-562083 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 08/Mar/21 00:55 Start Date: 08/Mar/21 00:55 Worklog Time Spent: 10m Work Description: github-actions[bot] commented on pull request #1778: URL: https://github.com/apache/hive/pull/1778#issuecomment-792393727 This pull request has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Feel free to reach out on the d...@hive.apache.org list if the patch is in need of reviews. This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 562083) Time Spent: 20m (was: 10m) > LLAP ShuffleHandler: upgrade to netty4 > -- > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Labels: pull-request-available > Time Spent: 20m > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is told to be a possible performance improvement compared to Netty3. > However, the refactor is not trivial, TEZ-4157 covers that more or less (the > code bases are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ > some other notes: Netty3 is EOL since 2016: > https://netty.io/news/2016/06/29/3-10-6-Final.html -- This message was sent by Atlassian Jira (v8.3.4#803005)
[jira] [Work logged] (HIVE-24524) LLAP ShuffleHandler: upgrade to netty4
[ https://issues.apache.org/jira/browse/HIVE-24524?focusedWorklogId=524006=com.atlassian.jira.plugin.system.issuetabpanels:worklog-tabpanel#worklog-524006 ] ASF GitHub Bot logged work on HIVE-24524: - Author: ASF GitHub Bot Created on: 14/Dec/20 17:40 Start Date: 14/Dec/20 17:40 Worklog Time Spent: 10m Work Description: abstractdog opened a new pull request #1778: URL: https://github.com/apache/hive/pull/1778 ### What changes were proposed in this pull request? ### Why are the changes needed? ### Does this PR introduce _any_ user-facing change? ### How was this patch tested? This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org Issue Time Tracking --- Worklog Id: (was: 524006) Remaining Estimate: 0h Time Spent: 10m > LLAP ShuffleHandler: upgrade to netty4 > -- > > Key: HIVE-24524 > URL: https://issues.apache.org/jira/browse/HIVE-24524 > Project: Hive > Issue Type: Improvement >Reporter: László Bodor >Assignee: László Bodor >Priority: Major > Time Spent: 10m > Remaining Estimate: 0h > > Tez already has a WIP patch for upgrading its shuffle handler to netty4. > Netty4 is told to be a possible performance improvement compared to Netty3. > However, the refactor is not trivial, TEZ-4157 covers that more or less (the > code bases are very similar). > Background: > netty4 migration guideline: > https://netty.io/wiki/new-and-noteworthy-in-4.0.html > articles of possible performance improvement: > https://blog.twitter.com/engineering/en_us/a/2013/netty-4-at-twitter-reduced-gc-overhead.html > https://developer.squareup.com/blog/upgrading-a-reverse-proxy-from-netty-3-to-4/ -- This message was sent by Atlassian Jira (v8.3.4#803005)
