If you are looking at alternatives check out the Infobright database. I'm in the process of switching my infrastructure over to that database, it uses considerably less memory and doesn't crash all the time. On Wed, Jul 7, 2010 at 8:17 PM, Chris wrote:

Hi Martin,

Well, nothing concrete yet, just putting together a typical, boring document to fling at management.

Even without clustering, MonetDB has beaten out the other DB options we've tried by a huge margin.

If I come across any specific performance concerns, I'll be sure to let you know.

Thanks, Chris

On 07/07/2010, at 4:19 PM, Martin Kersten wrote:

...

Chris wrote:

...

Hi, Hi Chris,

I'm currently evaluating MonetDB for business reporting, and looking for information on scalability, and I can't seem to find any information on clustering options.

Is there any support in MonetDB for clustering across multiple physical servers to improve performance?

Nope, that is not supported yet in the current releases. We look forward towards the requirements in terms of performance needed, e.g. schema/queries, which are hard to handle for the current version.

regards, Martin

...

Thanks, Chris

---

...

...

This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

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

...

This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Op 08-07-10 20:18, Mark Bucciarelli schreef:

Though I too am frustrated with the instability.

Typically what kind of things do you see regarding to instability? I'm currently running multiple webservices with a bunch of data and an open readonly SQL interface. The only crashes I observe are related: 1) silly SQL (thats basically why the webinterface is public, to catch those bugs) 2) usually some oddness with corrupt input (for example with copy into) Is your current production workload completely different? Because I have had uptimes exceeding > 4 months, and just to upgrade the software. Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.15 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEAREKAAYFAkw2Hq4ACgkQYH1+F2Rqwn3kLACfbUrjgRN1B0m6hqGqkI1uZ7jt b+YAnibEkctce6Ct8SD8/8sv60rTB6lf =jj+9 -----END PGP SIGNATURE-----

On Thu, Jul 8, 2010 at 3:05 PM, Mark Bucciarelli wrote:

On Thu, Jul 8, 2010 at 2:53 PM, Stefan de Konink wrote: -----BEGIN PGP SIGNED MESSAGE-----

Typically what kind of things do you see regarding to instability?

Other things I've noticed but haven't reported: mclient shell redirection looks strange to me. try this: $ mclient -l sql -d test -u test 1>1.out 2>2.out i've seen other unix utils that request a password navigate this successfully. i've seen behavior change depending on the case of the text in the SQL. i don't remember exactly what now, as i just worked around it. i've also had trouble with the server crashing when i tried setting ulimit's. sjoerd wrote great stuff, and i carefully followed it, but it didn't work for me. finally, from reading the code, i've seen that strdup() return values are often (usually?) ignored. there was even one case where fopen() return value was ignored, someone fixed that after I posted to the dev list. shared library handling makes some assumptions that aren't valid on all platforms. mx was segfaulting before sjoerd applied a variant of my patch (bug #2330). That's all I can think of at the moment. m

On Thu, Jul 08, 2010 at 03:05:35PM -0400, Mark Bucciarelli wrote:

On Thu, Jul 8, 2010 at 2:53 PM, Stefan de Konink wrote:

...

-----BEGIN PGP SIGNED MESSAGE-----

Typically what kind of things do you see regarding to instability?

Yesterday, it was wrong contraint id in fk violation error (bug #2598, open). Just fixed

Day before that, it was a select that reliably crashed db (#2583, which Niels fixed prompty, thank you).

Dropping a contraint corrupts DB (#2342). Reported back in March, with script that produces the behavior reliably, bug is still open.

Dropping a column corrupts DB (#2341). Same as above. Both were fixed some time ago and tests were added by sjoerd. These run fine, ie I closed them.

Niels

m

------------------------------------------------------------------------------ This SF.net email is sponsored by Sprint What will you do first with EVO, the first 4G phone? Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users

-- Niels Nes, Centrum Wiskunde & Informatica (CWI) Science Park 123, 1098 XG Amsterdam, The Netherlands room L3.14, phone ++31 20 592-4098 sip:4098@sip.cwi.nl url: http://www.cwi.nl/~niels e-mail: Niels.Nes@cwi.nl