[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread caneGuy
Github user caneGuy commented on the issue:

https://github.com/apache/spark/pull/18723
  
Thanks @kiszk @zsxwing 


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread zsxwing
Github user zsxwing commented on the issue:

https://github.com/apache/spark/pull/18723
  
Thanks! Merging to master.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/18723
  
Test PASSed.
Refer to this link for build results (access rights to CI server needed): 
https://amplab.cs.berkeley.edu/jenkins//job/SparkPullRequestBuilder/79943/
Test PASSed.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/18723
  
Merged build finished. Test PASSed.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/18723
  
**[Test build #79943 has 
finished](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/79943/testReport)**
 for PR 18723 at commit 
[`4c576a4`](https://github.com/apache/spark/commit/4c576a4d81c6af744fab28a9363061885a8aeb91).
 * This patch passes all tests.
 * This patch merges cleanly.
 * This patch adds no public classes.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread zsxwing
Github user zsxwing commented on the issue:

https://github.com/apache/spark/pull/18723
  
LGTM


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread SparkQA
Github user SparkQA commented on the issue:

https://github.com/apache/spark/pull/18723
  
**[Test build #79943 has 
started](https://amplab.cs.berkeley.edu/jenkins/job/SparkPullRequestBuilder/79943/testReport)**
 for PR 18723 at commit 
[`4c576a4`](https://github.com/apache/spark/commit/4c576a4d81c6af744fab28a9363061885a8aeb91).


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread zsxwing
Github user zsxwing commented on the issue:

https://github.com/apache/spark/pull/18723
  
ok to test


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread kiszk
Github user kiszk commented on the issue:

https://github.com/apache/spark/pull/18723
  
cc @JoshRosen


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread caneGuy
Github user caneGuy commented on the issue:

https://github.com/apache/spark/pull/18723
  
@kiszk Actually, i am confused with default value 16 too.
Yes, it occurs in specific scenario.In our case, it was large size block 
data which caused this issue.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread kiszk
Github user kiszk commented on the issue:

https://github.com/apache/spark/pull/18723
  
Thank you. 1. makes sense since memory is allocated by 2x at that point.
While I looked at a [PR](https://github.com/netty/netty/pull/464) for 
netty, I cannot understand why 16 was used.

I have another question to understand why it happens. Does this OOM occurs 
when any OpenBlocks message are received. Or, any specific scenario (e.g. 
receive a large message, a lot of multiple messages, or so on.)


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-25 Thread caneGuy
Github user caneGuy commented on the issue:

https://github.com/apache/spark/pull/18723
  
@kiszk Thanks for your time.
Yes,you are right.
1.OOM happens at io.netty.buffer.CompositeByteBuf.allocBuffer when one 
executor(client) fetch blocks from an other(server) , oom happens at server 
side before response to client.
2.Components number depends on chunks' length.Low-level is a list,since we 
do not need to remove or update component in the list, i can not think of any 
downside in our usage.However, if we do not consolidate we can avoid copy,which 
will improve the performance.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-24 Thread kiszk
Github user kiszk commented on the issue:

https://github.com/apache/spark/pull/18723
  
IIUC, this PR avoid data copy for consolidation in 
`netty.buffer.CompositeByteBuf.consolidateIfNeeded`. I think that it is 
reasonable.

I have two questions to understand this change.
1. Where OOM happens at `consoolidateIfNeeded` in the original code?
2. Is there any downside (e.g. performance degradation) to keep many 
components in `CompositeByteBuf`?




---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-24 Thread caneGuy
Github user caneGuy commented on the issue:

https://github.com/apache/spark/pull/18723
  
Can anyone help verify this?Thank too much.


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org



[GitHub] spark issue #18723: [SPARK-21517][CORE] Avoid copying memory when transfer c...

2017-07-24 Thread AmplabJenkins
Github user AmplabJenkins commented on the issue:

https://github.com/apache/spark/pull/18723
  
Can one of the admins verify this patch?


---
If your project is set up for it, you can reply to this email and have your
reply appear on GitHub as well. If your project does not have this feature
enabled and wishes so, or if the feature is enabled but not working, please
contact infrastructure at infrastruct...@apache.org or file a JIRA ticket
with INFRA.
---

-
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org