You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by Deepak Jaiswal <dj...@hortonworks.com> on 2017/03/20 21:59:01 UTC

Review Request 57794: HIVE-16260 : Remove parallel edges between semijoin and mpajoin

-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/57794/
-----------------------------------------------------------

Review request for hive and Jason Dere.


Bugs: HIVE-16260
    https://issues.apache.org/jira/browse/HIVE-16260


Repository: hive-git


Description
-------

Remove parallel edges between semijoin and mpajoin.
Added some extra config to tweak toe bloomfilter size and disable semijoin if bigtable is small.


Diffs
-----

  common/src/java/org/apache/hadoop/hive/conf/HiveConf.java 62908f9 
  ql/src/java/org/apache/hadoop/hive/ql/exec/vector/expressions/aggregates/VectorUDAFBloomFilter.java 2139eae 
  ql/src/java/org/apache/hadoop/hive/ql/exec/vector/expressions/aggregates/VectorUDAFBloomFilterMerge.java d2446d5 
  ql/src/java/org/apache/hadoop/hive/ql/optimizer/DynamicPartitionPruningOptimization.java c513ee5 
  ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java 47b229f 
  ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDAFBloomFilter.java 788aace 
  ql/src/test/queries/clientpositive/dynamic_partition_pruning.q d28da6e 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction.q 5482cdb 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_2.q 2306395 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_3.q 01650f8 
  ql/src/test/queries/clientpositive/mergejoin.q eecd105 
  ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction.q.out d32cb5c 


Diff: https://reviews.apache.org/r/57794/diff/1/


Testing
-------


Thanks,

Deepak Jaiswal


Re: Review Request 57794: HIVE-16260 : Remove parallel edges between semijoin and mpajoin

Posted by Jason Dere <jd...@hortonworks.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/57794/#review169582
-----------------------------------------------------------




common/src/java/org/apache/hadoop/hive/conf/HiveConf.java
Lines 2858 (patched)
<https://reviews.apache.org/r/57794/#comment241990>

    This setting is part of HIVE-16154 and should not be in this patch.



ql/src/java/org/apache/hadoop/hive/ql/exec/vector/expressions/aggregates/VectorUDAFBloomFilter.java
Line 78 (original), 80 (patched)
<https://reviews.apache.org/r/57794/#comment241994>

    I don't like that minEntries/scale factor changes have to be added in 3 places. Can this just be added to GenericUDAFBloomFilter, and modify getExpectedEntries() to incorporate the minEntries/factor logic?


- Jason Dere


On March 20, 2017, 9:59 p.m., Deepak Jaiswal wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/57794/
> -----------------------------------------------------------
> 
> (Updated March 20, 2017, 9:59 p.m.)
> 
> 
> Review request for hive and Jason Dere.
> 
> 
> Bugs: HIVE-16260
>     https://issues.apache.org/jira/browse/HIVE-16260
> 
> 
> Repository: hive-git
> 
> 
> Description
> -------
> 
> Remove parallel edges between semijoin and mpajoin.
> Added some extra config to tweak toe bloomfilter size and disable semijoin if bigtable is small.
> 
> 
> Diffs
> -----
> 
>   common/src/java/org/apache/hadoop/hive/conf/HiveConf.java 62908f9 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/expressions/aggregates/VectorUDAFBloomFilter.java 2139eae 
>   ql/src/java/org/apache/hadoop/hive/ql/exec/vector/expressions/aggregates/VectorUDAFBloomFilterMerge.java d2446d5 
>   ql/src/java/org/apache/hadoop/hive/ql/optimizer/DynamicPartitionPruningOptimization.java c513ee5 
>   ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java 47b229f 
>   ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDAFBloomFilter.java 788aace 
>   ql/src/test/queries/clientpositive/dynamic_partition_pruning.q d28da6e 
>   ql/src/test/queries/clientpositive/dynamic_semijoin_reduction.q 5482cdb 
>   ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_2.q 2306395 
>   ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_3.q 01650f8 
>   ql/src/test/queries/clientpositive/mergejoin.q eecd105 
>   ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction.q.out d32cb5c 
> 
> 
> Diff: https://reviews.apache.org/r/57794/diff/1/
> 
> 
> Testing
> -------
> 
> 
> Thanks,
> 
> Deepak Jaiswal
> 
>


Re: Review Request 57794: HIVE-16260 : Remove parallel edges between semijoin and mpajoin

Posted by Deepak Jaiswal <dj...@hortonworks.com>.
-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/57794/
-----------------------------------------------------------

(Updated March 21, 2017, 11:30 p.m.)


Review request for hive and Jason Dere.


Changes
-------

Implemented changes recommended.


Bugs: HIVE-16260
    https://issues.apache.org/jira/browse/HIVE-16260


Repository: hive-git


Description
-------

Remove parallel edges between semijoin and mpajoin.
Added some extra config to tweak toe bloomfilter size and disable semijoin if bigtable is small.


Diffs (updated)
-----

  common/src/java/org/apache/hadoop/hive/conf/HiveConf.java 62908f9 
  ql/src/java/org/apache/hadoop/hive/ql/optimizer/DynamicPartitionPruningOptimization.java c513ee5 
  ql/src/java/org/apache/hadoop/hive/ql/parse/TezCompiler.java 47b229f 
  ql/src/java/org/apache/hadoop/hive/ql/udf/generic/GenericUDAFBloomFilter.java 788aace 
  ql/src/test/queries/clientpositive/dynamic_partition_pruning.q d28da6e 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction.q 5482cdb 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_2.q 2306395 
  ql/src/test/queries/clientpositive/dynamic_semijoin_reduction_3.q 01650f8 
  ql/src/test/queries/clientpositive/mergejoin.q eecd105 
  ql/src/test/queries/clientpositive/vectorized_dynamic_partition_pruning.q d2ded71 
  ql/src/test/queries/clientpositive/vectorized_dynamic_semijoin_reduction.q 68b57ea 
  ql/src/test/queries/clientpositive/vectorized_dynamic_semijoin_reduction2.q be8e4af 
  ql/src/test/results/clientpositive/llap/dynamic_partition_pruning.q.out 96d998f 
  ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction.q.out d32cb5c 
  ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction_2.q.out 5f75977 
  ql/src/test/results/clientpositive/llap/dynamic_semijoin_reduction_3.q.out c272fc1 
  ql/src/test/results/clientpositive/llap/mergejoin.q.out ae99e66 
  ql/src/test/results/clientpositive/llap/vectorized_dynamic_partition_pruning.q.out 0182a46 
  ql/src/test/results/clientpositive/llap/vectorized_dynamic_semijoin_reduction.q.out e62d913 
  ql/src/test/results/clientpositive/llap/vectorized_dynamic_semijoin_reduction2.q.out d9fd706 
  ql/src/test/results/clientpositive/tez/explainanalyze_3.q.out 32609eb 
  ql/src/test/results/clientpositive/tez/explainuser_3.q.out da52b0a 


Diff: https://reviews.apache.org/r/57794/diff/2/

Changes: https://reviews.apache.org/r/57794/diff/1-2/


Testing
-------


Thanks,

Deepak Jaiswal