[
https://issues.apache.org/jira/browse/IMPALA-7655?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16659482#comment-16659482
]
Paul Rogers edited comment on IMPALA-7655 at 10/23/18 6:42 PM:
---------------------------------------------------------------
To summarize the proposed changes:
* Rewrite {{isnull()}} to use {{CASE}}. (Done)
* Rewrite {{coalesce()}} to use {{CASE}}. (Done)
* Rewrite {{if()}} to use {{CASE}}.
* Split conditional rewrite tests out of {{ExprRewriteRulesTest}} into a new
test class: {{SimplifyConditionalRulesTest}}. (Done)
* Modify the above to test the new rewrite rules.
* Remove {{conditional-functions.[h|cc]}}
* Add tests for the additional simplifications described above. (Done)
* Run BE tests to ensure that query functionality is unchanged.
* Run ad-hoc performance tests, of the kind the ticket description, to check
resulting performance.
was (Author: paul.rogers):
To summarize the proposed changes:
* Rewrite {{ifnull()}} and {{nvl()}} to {{isnull()}}. (Done)
* Rewrite {{isnull()}} to use {{CASE}}. (Done)
* Rewrite {{coalesce()}} to use {{CASE}}. (Done)
* Rewrite {{if()}} to use {{CASE}}.
* Modify the {{decode()}} rewrite to use {{IS NOT DISTINCT}}. (Done.)
* Split conditional rewrite tests out of {{ExprRewriteRulesTest}} into a new
test class: {{SimplifyConditionalRulesTest}}. (Done)
* Implement special handling for {{decode()}} in the builtins table, then move
code for rewrite out of {{CaseExpr}} and into {{SimplifyConditionalRules}}.
(Done.)
* Address IMPALA-7741 as this is needed to allow removing BE implementations
of functions while keeping the functions in the FE function registry. (Done.)
* Remove {{conditional-functions.[h|cc]}}
* Change handling of {{nvl2()}} and {{nullif()}} to perform the rewriting in
{{SimplifyConditionalRules}} rather than when creating the function expression.
And the simplifications suggested above.
* Modify the above to test for the new rewrite rules.
* No tests exist to verify the rewrite of {{decode()}} into {{CASE}}. Add such
tests, including the use of the {{IS NOT DISTINCT}} operator as noted above.
(Done)
* Add tests for the additional simplifications described above. (Done)
* Run BE tests to ensure that query functionality is unchanged.
* Run ad-hoc performance tests, of the kind the ticket description, to check
resulting performance.
> Codegen output for conditional functions (if,isnull, coalesce) is very
> suboptimal
> ---------------------------------------------------------------------------------
>
> Key: IMPALA-7655
> URL: https://issues.apache.org/jira/browse/IMPALA-7655
> Project: IMPALA
> Issue Type: Improvement
> Components: Backend
> Reporter: Tim Armstrong
> Priority: Major
> Labels: codegen, perf, performance
>
> https://gerrit.cloudera.org/#/c/11565/ provided a clue that an aggregation
> involving an if() function was very slow, 10x slower than the equivalent
> version using a case:
> {noformat}
> [localhost:21000] default> set num_nodes=1; set mt_dop=1; select count(case
> when l_orderkey is NULL then 1 else NULL end) from
> tpch10_parquet.lineitem;summary;
> NUM_NODES set to 1
> MT_DOP set to 1
> Query: select count(case when l_orderkey is NULL then 1 else NULL end) from
> tpch10_parquet.lineitem
> Query submitted at: 2018-10-04 11:17:31 (Coordinator:
> http://tarmstrong-box:25000)
> Query progress can be monitored at:
> http://tarmstrong-box:25000/query_plan?query_id=274b2a6f35cefe31:95a1964200000000
> +----------------------------------------------------------+
> | count(case when l_orderkey is null then 1 else null end) |
> +----------------------------------------------------------+
> | 0 |
> +----------------------------------------------------------+
> Fetched 1 row(s) in 0.51s
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> | Operator | #Hosts | Avg Time | Max Time | #Rows | Est. #Rows | Peak
> Mem | Est. Peak Mem | Detail |
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> | 01:AGGREGATE | 1 | 44.03ms | 44.03ms | 1 | 1 | 25.00
> KB | 10.00 MB | FINALIZE |
> | 00:SCAN HDFS | 1 | 411.57ms | 411.57ms | 59.99M | -1 | 16.61
> MB | 88.00 MB | tpch10_parquet.lineitem |
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> [localhost:21000] default> set num_nodes=1; set mt_dop=1; select
> count(if(l_orderkey is NULL, 1, NULL)) from tpch10_parquet.lineitem;summary;
> NUM_NODES set to 1
> MT_DOP set to 1
> Query: select count(if(l_orderkey is NULL, 1, NULL)) from
> tpch10_parquet.lineitem
> Query submitted at: 2018-10-04 11:23:07 (Coordinator:
> http://tarmstrong-box:25000)
> Query progress can be monitored at:
> http://tarmstrong-box:25000/query_plan?query_id=8e46ab1b84c4dbff:2786ca2600000000
> +----------------------------------------+
> | count(if(l_orderkey is null, 1, null)) |
> +----------------------------------------+
> | 0 |
> +----------------------------------------+
> Fetched 1 row(s) in 1.01s
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> | Operator | #Hosts | Avg Time | Max Time | #Rows | Est. #Rows | Peak
> Mem | Est. Peak Mem | Detail |
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> | 01:AGGREGATE | 1 | 422.07ms | 422.07ms | 1 | 1 | 25.00
> KB | 10.00 MB | FINALIZE |
> | 00:SCAN HDFS | 1 | 511.13ms | 511.13ms | 59.99M | -1 | 16.61
> MB | 88.00 MB | tpch10_parquet.lineitem |
> +--------------+--------+----------+----------+--------+------------+----------+---------------+-------------------------+
> {noformat}
> It turns out that this is because we don't have good codegen support for
> ConditionalFunction, and just fall back to emitting a call to the interpreted
> path:
> https://github.com/apache/impala/blob/master/be/src/exprs/conditional-functions.cc#L28
> See CaseExpr for an example of much better codegen support:
> https://github.com/apache/impala/blob/master/be/src/exprs/case-expr.cc#L178
--
This message was sent by Atlassian JIRA
(v7.6.3#76005)
---------------------------------------------------------------------
To unsubscribe, e-mail: issues-all-unsubscr...@impala.apache.org
For additional commands, e-mail: issues-all-h...@impala.apache.org
