phoenix-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "James Taylor (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (PHOENIX-2126) Improving performance of merge sort by multi-threaded and minheap implementation
Date Fri, 24 Jul 2015 01:30:04 GMT

    [ https://issues.apache.org/jira/browse/PHOENIX-2126?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=14639791#comment-14639791
] 

James Taylor commented on PHOENIX-2126:
---------------------------------------

Nice work, [~ankit.singhal]. Would it make sense to use the same thread pool that we use for
querying so that the thread pools are not competing with each other?

I think the perf gain is from both parallelizing and using a heap. The heap helps avoid checking
every Tuple at the head of each iterator again and again to find the largest.

> Improving performance of merge sort by multi-threaded and minheap implementation
> --------------------------------------------------------------------------------
>
>                 Key: PHOENIX-2126
>                 URL: https://issues.apache.org/jira/browse/PHOENIX-2126
>             Project: Phoenix
>          Issue Type: Improvement
>    Affects Versions: 4.1.0, 4.2.0
>            Reporter: Ankit Singhal
>         Attachments: PHOENIX-2126_v1.0.patch
>
>
> {code}
> CREATE TABLE IF NOT EXISTS test (
> dim1 INTEGER NOT NULL,
> A.B INTEGER,
> A.M DECIMAL,
> CONSTRAINT PK PRIMARY KEY
> (dim1))
> SALT_BUCKETS =256,DEFAULT_COLUMN_FAMILY='A';
> {code}
> *Query to benchmark:-*
> {code}
> select dim1,sum(b),sum(m) from test where Datemth>=201505 and Datemth<=201505 and
dim1 IS NOT NULL  group by dim1 order by sum(m) desc nulls last limit 10;
> {code}
> *current scenario:-*
> *CASE 1: * consider the case when dim1 is high cardinality attribute (10K+) and table
have salt bucket set to 256, we will get 256 iterators from above query at the client and
MergeSortRowKeyResultIterator has to merge these 256 iterators with single thread. So let's
say each iterator has 10k tuples returned, then merge sort needs to merge 2.5M tuples which
will be costly if it is done with single thread and the query spend most of its time on client
> *CASE 2: * consider the case when dim1 is high cardinality attribute (10K+) and table
have salt bucket set to 1, we will get 1 iterator from  above query at the client and MergeSortRowKeyResultIterator
doesn't need to merge anything. Here, it is fine with single thread.
> *CASE 3: * consider the case when dim1 is low cardinality attribute (10-100) and table
have salt bucket set to 256, we will get 256 iterator from  above query at the client and
MergeSortRowKeyResultIterator has to merge these 256 iterators with single thread. here the
single thread is also fine as he has to merge only 2560 tuples.
> *Solution for case1 problem is:-*
> Optimized the implementation of merging 'n'-sorted iterators(having 'm' tuples)  by using
"min heap" which optimizes the time complexity from 
> O(n2m) to O(nmLogn) (as heapify takes (Logn) time).
> And, By using multiple-threads('t') to process group of iterators which further optimized
the complexity to 
> T(nm)=T(nm)/t+T(t)



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Mime
View raw message