You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by Mahsa Mofidpoor <mo...@gmail.com> on 2012/05/23 02:11:00 UTC

Using indexes for predicates does not increase performance.

Hi,

I did a simple experiment to see the performance difference between normal
execution of a query and the index-based execution. However, the response
time in non-indexed approach is astonishingly smaller. Is it because of the
small data size or the single setup node I use?
I appreciate any help.

Mahsa

Ubuntu 10.0.4
Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
RAM 4GB

Hadoop 0.20.2 (*Single setup node*)
Hive-0.9.0
table size ranges from 4MB to 100MB

hive.optimize.index.filter=true
query: select * from table1 where col1='Mark';

RE: Using indexes for predicates does not increase performance.

Posted by Chinna Rao Lalam <ch...@huawei.com>.
Hi,

Try with bigger data.....Then u can see the performace difference...

Hope It Helps,
Chinna Rao Lalam
________________________________________
From: Mahsa Mofidpoor [mofidpoor@gmail.com]
Sent: Wednesday, May 23, 2012 8:11 AM
To: dev@hive.apache.org
Subject: Using indexes for predicates does not increase performance.

Hi,

I did a simple experiment to see the performance difference between normal
execution of a query and the index-based execution. However, the response
time in non-indexed approach is astonishingly smaller. Is it because of the
small data size or the single setup node I use?
I appreciate any help.

Mahsa

Ubuntu 10.0.4
Intel(R) Core(TM) i5-2400 CPU @ 3.10GHz
RAM 4GB

Hadoop 0.20.2 (*Single setup node*)
Hive-0.9.0
table size ranges from 4MB to 100MB

hive.optimize.index.filter=true
query: select * from table1 where col1='Mark';