[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-09-27 Thread Dmitriy V. Ryaboy (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13465341#comment-13465341 ] Dmitriy V. Ryaboy commented on PIG-483: --- I think we all like Bill's proposal, but would

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-09-27 Thread Jie Li (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13465334#comment-13465334 ] Jie Li commented on PIG-483: Have we agree on this approach yet? I like Bill's proposal but I'm a

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-09-27 Thread Dmitriy V. Ryaboy (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13465280#comment-13465280 ] Dmitriy V. Ryaboy commented on PIG-483: --- bump -- Jie, any chance you have some time to

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-09-05 Thread Dmitriy V. Ryaboy (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13448536#comment-13448536 ] Dmitriy V. Ryaboy commented on PIG-483: --- I think Bill's suggestion is cleaner, but in t

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-08-01 Thread Bill Graham (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13426749#comment-13426749 ] Bill Graham commented on PIG-483: - Instead of modifying the job graph at runtime (in this cas

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-06-28 Thread Jie Li (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13403635#comment-13403635 ] Jie Li commented on PIG-483: Ooops, forgot that the sample job will always use 1 reducer instead

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-06-27 Thread Jie Li (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13402661#comment-13402661 ] Jie Li commented on PIG-483: For the skew join, if the partition table turns out to be small, the

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-06-25 Thread Jie Li (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13401017#comment-13401017 ] Jie Li commented on PIG-483: Thanks Bill! Maybe we can implement a SkipJob for the moment? I can

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-06-25 Thread Bill Graham (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13400602#comment-13400602 ] Bill Graham commented on PIG-483: - Good point. In the current design the job graph is assumed

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-06-22 Thread Jie Li (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13399593#comment-13399593 ] Jie Li commented on PIG-483: As Dmitriy pointed out, we need to optimize this at runtime. For ord

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2012-04-29 Thread Dmitriy V. Ryaboy (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13264635#comment-13264635 ] Dmitriy V. Ryaboy commented on PIG-483: --- Note that since parallelism can be determined

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2011-04-08 Thread Daniel Dai (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13017668#comment-13017668 ] Daniel Dai commented on PIG-483: No, the number of reducers are currently determined before y

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2011-04-08 Thread Zubair Nabi (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13017479#comment-13017479 ] Zubair Nabi commented on PIG-483: - But how can one make the call that the data is small enoug

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2011-04-04 Thread Daniel Dai (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13015511#comment-13015511 ] Daniel Dai commented on PIG-483: Here the focus is one single reduce not the size of data. Cu

[jira] [Commented] (PIG-483) PERFORMANCE: different strategies for large and small order bys

2011-04-04 Thread Zubair Nabi (JIRA)
[ https://issues.apache.org/jira/browse/PIG-483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13015390#comment-13015390 ] Zubair Nabi commented on PIG-483: - Hi, I would like more information about this project. To