[ 
https://issues.apache.org/jira/browse/KUDU-2483?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16521503#comment-16521503
 ] 

jin xing commented on KUDU-2483:
--------------------------------

Sorry for late reply. True, pushing down bloom filters for scan is quite 
useful. `Join` is a typical scenario. But this feature I think is not 
necessarily provided for join.
Spark currently doesn't have support for push-down bloom filters. We have 
implement this feature and targeting to contribute to both Spark and Kudu.
And I also agrree that we should go by exposing a bloom filter from Kudu.

Basically our work contains below parts:
1. A Java version Bloom filter.
2. `BroadcastBloomFilterHashJoinExec` -- it can generate bloom filters from the 
small table and send the BF to Kudu client.
3. Kudu RDD accepts the BFs generated by `BroadcastBloomFilterHashJoinExec` and 
merge as Predicate and send to TServer.
4. TServer accepts the BFs and send back filtered data

For Part-2, it will be a patch to Spark and other parts are implemented within 
Kudu scope.

If the implementation is interested. I can created some subtasks(this JIRA will 
be an umbrella) and submit the patch.

> Scan tablets with bloom filter
> ------------------------------
>
>                 Key: KUDU-2483
>                 URL: https://issues.apache.org/jira/browse/KUDU-2483
>             Project: Kudu
>          Issue Type: New Feature
>          Components: client
>            Reporter: jin xing
>            Priority: Major
>
> Join is really common/popular in Spark SQL, in this JIRA I take broadcast 
> join as an example and describe how Kudu's bloom filter can help accelerate 
> distributed computing.
> Spark runs broadcast join with below steps:
> 1. When do broadcast join, we have a small table and a big table; Spark will 
> read all data from small table to one worker and build a hash table;
> 2. The generated hash table from step 1 is broadcasted to all the workers, 
> which will read the splits from big table;
> 3. Workers start fetching and iterating all the splits of big table and see 
> if the joining keys exists in the hash table; Only matched joining keys is 
> retained.
> From above, step 3 is the heaviest, especially when the worker and split 
> storage is not on the same host and bandwith is limited. Actually the cost 
> brought by step 3 is not always necessary. Think about below scenario:
> {code:none}
> Small table A
> id      name
> 1      Jin
> 6      Xing
> Big table B
> id     age
> 1      10
> 2      21
> 3      33
> 4      65
> 5      32
> 6      23
> 7      18
> 8      20
> 9      22
> {code}
> Run query with SQL: *select * from A inner join B on A.id=B.id*
> It's pretty straight that we don't need to fetch all the data from Table B, 
> because the number of matched keys is really small;
> I propose to use small table to build a bloom filter(BF) and use the 
> generated BF as a predicate/filter to fetch data from big table, thus:
> 1. Much traffic/bandwith is saved.
> 2. Less data to processe by worker
> Broadcast join is just an example, other types of join will also benefit if 
> we scan with a BF
> In a nutshell, I think Kudu can provide an iterface, by which user can scan 
> data with bloom filters



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)

Reply via email to