[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14001506#comment-14001506 ] Lefty Leverenz commented on HIVE-4440: -- Documented *hive.smbjoin.cache.rows* in the wiki, and revised *hive.mapjoin.bucket.cache.size*: * [hive.smbjoin.cache.rows | https://cwiki.apache.org/confluence/display/Hive/Configuration+Properties#ConfigurationProperties-hive.smbjoin.cache.rows] * [hive.mapjoin.bucket.cache.size | https://cwiki.apache.org/confluence/display/Hive/Configuration+Properties#ConfigurationProperties-hive.mapjoin.bucket.cache.size] SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Fix For: 0.12.0 Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14001270#comment-14001270 ] Lefty Leverenz commented on HIVE-4440: -- Hive 0.13.0 did not remove *hive.mapjoin.bucket.cache.size*. Also, the comment that says it should be removed has a typo in the name of the new parameter -- it should be *hive.smbjoin.cache.rows*, not hive.smbjoin.cache.row: {quote} +// hive.mapjoin.bucket.cache.size has been replaced by hive.smbjoin.cache.row, +// need to remove by hive .13. Also, do not change default (see SMB operator) {quote} Instead of creating a new jira for this, I'll add a comment on HIVE-6586 (for HIVE-6037). SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Fix For: 0.12.0 Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14001280#comment-14001280 ] Lefty Leverenz commented on HIVE-4440: -- Here's the comment I added to HIVE-6586: * [comment about hive.mapjoin.bucket.cache.size and hive.smbjoin.cache.rows | https://issues.apache.org/jira/browse/HIVE-6586?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=14001274#comment-14001274] SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Fix For: 0.12.0 Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message was sent by Atlassian JIRA (v6.2#6252)
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13659362#comment-13659362 ] Hudson commented on HIVE-4440: -- Integrated in Hive-trunk-h0.21 #2105 (See [https://builds.apache.org/job/Hive-trunk-h0.21/2105/]) HIVE-4440 SMB Operator spills to disk like it's 1999 (Gunther Hagleitner via omalley) (Revision 1483084) Result = FAILURE omalley : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVNview=revrev=1483084 Files : * /hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java * /hive/trunk/conf/hive-default.xml.template * /hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/SMBMapJoinOperator.java SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Fix For: 0.12.0 Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13659934#comment-13659934 ] Hudson commented on HIVE-4440: -- Integrated in Hive-trunk-hadoop2 #199 (See [https://builds.apache.org/job/Hive-trunk-hadoop2/199/]) HIVE-4440 SMB Operator spills to disk like it's 1999 (Gunther Hagleitner via omalley) (Revision 1483084) Result = FAILURE omalley : http://svn.apache.org/viewcvs.cgi/?root=Apache-SVNview=revrev=1483084 Files : * /hive/trunk/common/src/java/org/apache/hadoop/hive/conf/HiveConf.java * /hive/trunk/conf/hive-default.xml.template * /hive/trunk/ql/src/java/org/apache/hadoop/hive/ql/exec/SMBMapJoinOperator.java SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Fix For: 0.12.0 Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13658727#comment-13658727 ] Gunther Hagleitner commented on HIVE-4440: -- [~owen.omalley]: Ran all tests - no failures. SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13657645#comment-13657645 ] Owen O'Malley commented on HIVE-4440: - Patch.2 looks good. I agree with Namit that supporting the old config makes sense. Can you run tests on it. We might consider warning in Hive 0.13 and removing in Hive 0.14. *smile* SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13647395#comment-13647395 ] Gunther Hagleitner commented on HIVE-4440: -- Thanks :-) Patch .2 honors the old parameter unless it's at the default in which case it uses the new one. I also put documentation around it. You bring up a good point, but are you sure it's necessary to support both in this case though? It's just slightly ugly in the code and requires us to move in again to remove later. My thinking is this: If you use the old parameter, it's probably because you needed to up it to get better performance - in this case the new default should most likely be ok for you. Do you think there's going to be cases where this falls flat? SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Attachments: HIVE-4440.1.patch, HIVE-4440.2.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira
[jira] [Commented] (HIVE-4440) SMB Operator spills to disk like it's 1999
[ https://issues.apache.org/jira/browse/HIVE-4440?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanelfocusedCommentId=13645178#comment-13645178 ] Namit Jain commented on HIVE-4440: -- I really like the title of the jira. Changing the parameter name is backward incompatible. Can you support both the current parameter and the proposed parameter for now ? Document it clearly, and say that the current parameter hive.mapjoin.bucket.cache.size will not be supported for this from 0.13 or something like that. SMB Operator spills to disk like it's 1999 -- Key: HIVE-4440 URL: https://issues.apache.org/jira/browse/HIVE-4440 Project: Hive Issue Type: Bug Reporter: Gunther Hagleitner Assignee: Gunther Hagleitner Attachments: HIVE-4440.1.patch I was recently looking into some performance issue with a query that used SMB join and was running really slow. Turns out that the SMB join by default caches only 100 values per key before spilling to disk. That seems overly conservative to me. Changing the parameter resulted in a ~5x speedup - quite significant. The parameter is: hive.mapjoin.bucket.cache.size Which right now is only used the SMB Operator as far as I can tell. The parameter was introduced originally (3 yrs ago) for the map join operator (looks like pre-SMB) and set to 100 to avoid OOM. That seems to have been in a different context though where you had to avoid running out of memory with the cached hash table in the same process, I think. Two things I'd like to propose: a) Rename it to what it does: hive.smbjoin.cache.rows b) Set it to something less restrictive: 1 If you string together a 5 table smb join with a map join and a map-side group by aggregation you might still run out of memory, but the renamed parameter should be easier to find and reduce. For most queries, I would think that 1 is still a reasonable number to cache (On the reduce side we use 25000 for shuffle joins). -- This message is automatically generated by JIRA. If you think it was sent incorrectly, please contact your JIRA administrators For more information on JIRA, see: http://www.atlassian.com/software/jira