[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-12-08 Thread zhijiang (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16991107#comment-16991107 ] zhijiang commented on FLINK-14845: -- [~pnowojski] Yes, exactly we would verify the effects via benchmark

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-12-08 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16990786#comment-16990786 ] Piotr Nowojski commented on FLINK-14845: Thanks [~kevin.cyj] and [~zjwang] for your efforts!

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-12-07 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16990697#comment-16990697 ] Yingjie Cao commented on FLINK-14845: - Fix via 66d4d7da2d8b717f420509d9785fad0880562f10 on master.

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-12-01 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16985873#comment-16985873 ] Yingjie Cao commented on FLINK-14845: - Based on the above discussions, I have given an

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-27 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16984097#comment-16984097 ] Yingjie Cao commented on FLINK-14845: - > So I think it's fine to not make it pluggable in the first

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-27 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16983552#comment-16983552 ] Piotr Nowojski commented on FLINK-14845: Regarding the org.lz4:lz4-java dependency. This one is

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-27 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16983338#comment-16983338 ] Yingjie Cao commented on FLINK-14845: - > Would it be more complicate to request a buffer from the

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-26 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16982446#comment-16982446 ] Piotr Nowojski commented on FLINK-14845: For the output, writing back the compressed bytes in

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-26 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16982271#comment-16982271 ] Yingjie Cao commented on FLINK-14845: - > Where are you going to compress data into? Each compressor

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-23 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16980729#comment-16980729 ] Piotr Nowojski commented on FLINK-14845: Maybe before going further I have a couple of

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-21 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16979871#comment-16979871 ] Yingjie Cao commented on FLINK-14845: - [~pnowojski] I totally agree with you. Then I will work on

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-21 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16979249#comment-16979249 ] Piotr Nowojski commented on FLINK-14845: [~kevin.cyj], I think I would prefer the 3rd option,

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16979002#comment-16979002 ] Yingjie Cao commented on FLINK-14845: - [~pnowojski] , thanks for proposing these options. #

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Jingsong Lee (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978343#comment-16978343 ] Jingsong Lee commented on FLINK-14845: -- > doesn't Blink work on batches of records? Couldn't the

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978325#comment-16978325 ] Piotr Nowojski commented on FLINK-14845: Re, [~lzljs3620320]: doesn't Blink work on batches of

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978297#comment-16978297 ] Chesnay Schepler commented on FLINK-14845: -- [~kevin.cyj] Cluster partitions may also be read by

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978288#comment-16978288 ] Yingjie Cao commented on FLINK-14845: - [~chesnay] A choice maybe make the compression codec cluster

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978284#comment-16978284 ] Yingjie Cao commented on FLINK-14845: - [~pnowojski]  You are right, taking both Blocking and

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Chesnay Schepler (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978268#comment-16978268 ] Chesnay Schepler commented on FLINK-14845: -- How would this work with cluster partitions?

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Jingsong Lee (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978257#comment-16978257 ] Jingsong Lee commented on FLINK-14845: -- [~pnowojski] put forward a good thought for the future. I

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Piotr Nowojski (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978243#comment-16978243 ] Piotr Nowojski commented on FLINK-14845: +1 for the future. Could be useful also for

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978241#comment-16978241 ] Yingjie Cao commented on FLINK-14845: - Doing compression and decompression with task threads may be

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Stephan Ewen (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978203#comment-16978203 ] Stephan Ewen commented on FLINK-14845: -- For combining compression and low latency, we need to see

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978184#comment-16978184 ] Yingjie Cao commented on FLINK-14845: - [~lzljs3620320] For small data, for example, dozens of byte,

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-20 Thread Jingsong Lee (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978175#comment-16978175 ] Jingsong Lee commented on FLINK-14845: -- Thanks [~kevin.cyj] A little confuse about point 8, Are

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-19 Thread Yingjie Cao (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978153#comment-16978153 ] Yingjie Cao commented on FLINK-14845: - To verify how many improvements we could gain from

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-19 Thread Jingsong Lee (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16978019#comment-16978019 ] Jingsong Lee commented on FLINK-14845: -- [~sewen] Sorry for misunderstanding. What I want to say is

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-19 Thread Kurt Young (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16977982#comment-16977982 ] Kurt Young commented on FLINK-14845: [~sewen] Currently we set all shuffle type to `BLOCKING` in

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-19 Thread Stephan Ewen (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16977693#comment-16977693 ] Stephan Ewen commented on FLINK-14845: -- [~lzljs3620320] Quick question for clarification: In the

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-19 Thread Stephan Ewen (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16977441#comment-16977441 ] Stephan Ewen commented on FLINK-14845: -- Kurt and me had a quick chat about this. On the receiver

[jira] [Commented] (FLINK-14845) Introduce data compression to blocking shuffle.

2019-11-18 Thread Jingsong Lee (Jira)
[ https://issues.apache.org/jira/browse/FLINK-14845?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel=16977096#comment-16977096 ] Jingsong Lee commented on FLINK-14845: -- Big +1 for this feature. There will be a lot of shuffles