You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@hive.apache.org by Namit Jain <nj...@fb.com> on 2013/01/31 05:23:10 UTC

reduced unit test timings

I have noticed that the time taken to run the unit tests has reduced considerably (it has become nearly half) from the last week or so.
Just wondering, if anyone else has noticed this too.

If yes, does anyone know the root cause of this speedup ?

Thanks,
-namit

Re: reduced unit test timings

Posted by Namit Jain <nj...@fb.com>.
I run tests on a parallel cluster (8 machines).
For that, the test time has gone down from 2:15hours to approx. 1:15


On 1/31/13 11:55 AM, "Ashutosh Chauhan" <ha...@apache.org> wrote:

>I am not sure about half, but
>https://issues.apache.org/jira/browse/HIVE-3947 has certainly helped. Both
>MiniMRCliDriver and NegativeMiniMRCliDriver used to remain in hung state
>for ~10 minutes after all tests have run and minicluster is tearing down.
>That patch has saved atleast ~15 mins for test runs in my environment.
>
>Thanks to Navis for that!
>
>Ashutosh
>
>
>On Wed, Jan 30, 2013 at 8:23 PM, Namit Jain <nj...@fb.com> wrote:
>
>> I have noticed that the time taken to run the unit tests has reduced
>> considerably (it has become nearly half) from the last week or so.
>> Just wondering, if anyone else has noticed this too.
>>
>> If yes, does anyone know the root cause of this speedup ?
>>
>> Thanks,
>> -namit
>>

Re: reduced unit test timings

Posted by Ashutosh Chauhan <ha...@apache.org>.
I am not sure about half, but
https://issues.apache.org/jira/browse/HIVE-3947 has certainly helped. Both
MiniMRCliDriver and NegativeMiniMRCliDriver used to remain in hung state
for ~10 minutes after all tests have run and minicluster is tearing down.
That patch has saved atleast ~15 mins for test runs in my environment.

Thanks to Navis for that!

Ashutosh


On Wed, Jan 30, 2013 at 8:23 PM, Namit Jain <nj...@fb.com> wrote:

> I have noticed that the time taken to run the unit tests has reduced
> considerably (it has become nearly half) from the last week or so.
> Just wondering, if anyone else has noticed this too.
>
> If yes, does anyone know the root cause of this speedup ?
>
> Thanks,
> -namit
>