Re: [MonetDB-users] Problem with optimizer ?

On Fri, May 13, 2011 at 06:15:02PM +0300, Georgios Garmpis wrote:

Both PLAN and EXPLAIN have the same result (being killed by OS' OOM killer). So, does size of data mater?

In general it does for the resource consumption during query execution.

Actually my tables are emtpy.

This informtation, and fact that your query appears to make the make the server grows to an extend that it get's killed by the OS' OOM killer are curcial information for us. As far as I can tell now, this suggests that already the query transtlation or optimization failes, i.e., is does not even get to executing the generated plan. Could you please file a bug report via http://bugs.monetdb.org/ including all informtation required to reporduce the crash? Thanks! Stefan

On Fri, May 13, 2011 at 2:57 PM, Stefan Manegold wrote:

...

On Fri, May 13, 2011 at 01:27:13PM +0200, Fabian Groffen wrote:

...

On 13-05-2011 12:54:05 +0300, Georgios Garmpis wrote:

...

To be more clear,

I cannot figure out the reason why mserver acts like that.

Query doesn't seem very complicated to me. Should i refine my query or is it a problem of mserver (e.g. configuration issue, bug etc)?

Try prefixing EXPLAIN to your "simple" query, and you'll get an idea what mserver is doing for you to perform the query.

Indeed.

Maybe you could prefix your query with both PLAN and EXPLAIN (one at a time) and share both results with us to see how MonetDB translates your query.

Also: how large is your data, i.e., how many records are there in the tables involved, and how large (number of records) is the expected result of your query?

Stefan

...

------------------------------------------------------------------------------

...

Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

-- | Stefan.Manegold @ CWI.nl | DB Architectures (INS1) | | http://CWI.nl/~manegold/ | Science Park 123 (L321) | | Tel.: +31 (0)20 592-4212 | 1098 XG Amsterdam (NL) |

------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

-- George Garbis

------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay

_______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

-- | Stefan.Manegold @ CWI.nl | DB Architectures (INS1) | | http://CWI.nl/~manegold/ | Science Park 123 (L321) | | Tel.: +31 (0)20 592-4212 | 1098 XG Amsterdam (NL) |