[
https://issues.apache.org/jira/browse/KUDU-1683?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16201639#comment-16201639
]
HeLifu commented on KUDU-1683:
------------------------------
the front sql gets up to 20x speedup after pushing runtime filters to
kudu-tserver(bloom filter).
{code:java}
[hadoop956.hz.163.org:21000] > select c.* from orders o join customer c on
c.c_custkey = o.o_custkey where o_orderkey = 1125;
Query: select c.* from orders o join customer c on c.c_custkey = o.o_custkey
where o_orderkey = 1125
Query submitted at: 2017-10-12 11:11:50 (Coordinator:
http://hadoop956.hz.163.org:25000)
Query progress can be monitored at:
http://hadoop956.hz.163.org:25000/query_plan?query_id=da4cb4fbc1da9ca8:ad2baf1c00000000
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
| c_custkey | c_name | c_address | c_nationkey | c_phone
| c_acctbal | c_mktsegment | c_comment
|
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
| 24123473 | Customer#024123473 | JBUCtxOA8p5cT,u3 | 23 |
33-270-234-9550 | -567.49 | HOUSEHOLD | onic pinto beans haggle about the
ironic, bold foxes. carefully special asymptotes use requests. re |
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
Fetched 1 row(s) in 43.08s
[hadoop953.hz.163.org:21000] > select c.* from orders o join customer c on
c.c_custkey = o.o_custkey where o_orderkey = 1125;
Query: select c.* from orders o join customer c on c.c_custkey = o.o_custkey
where o_orderkey = 1125
Query submitted at: 2017-10-12 10:59:26 (Coordinator:
http://hadoop953.hz.163.org:25000)
Query progress can be monitored at:
http://hadoop953.hz.163.org:25000/query_plan?query_id=d84a9b136e7d3123:55edbdf300000000
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
| c_custkey | c_name | c_address | c_nationkey | c_phone
| c_acctbal | c_mktsegment | c_comment
|
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
| 24123473 | Customer#024123473 | JBUCtxOA8p5cT,u3 | 23 |
33-270-234-9550 | -567.49 | HOUSEHOLD | onic pinto beans haggle about the
ironic, bold foxes. carefully special asymptotes use requests. re |
+-----------+--------------------+------------------+-------------+-----------------+-----------+--------------+-----------------------------------------------------------------------------------------------------+
Fetched 1 row(s) in 2.04s
{code}
> Kudu client support for pushing runtime min/max filters
> -------------------------------------------------------
>
> Key: KUDU-1683
> URL: https://issues.apache.org/jira/browse/KUDU-1683
> Project: Kudu
> Issue Type: New Feature
> Components: client, perf
> Affects Versions: 1.0.0
> Reporter: Matthew Jacobs
> Labels: impala
>
> Impala would like to generate runtime min/max filters to be pushed to Kudu,
> at least for scan tokens that haven't been opened yet.
> https://issues.cloudera.org/browse/IMPALA-4252
--
This message was sent by Atlassian JIRA
(v6.4.14#64029)
