[MonetDB-users] PHP Benchmarks of Native versus Cimpl?
Hi~ I've recently converted our monet install from the cimpl version of the php interface to the new native socket interface. I see the readme references some benchmarks at a url that does not appear to exist (http://www.ins.cwi.nl/~gm/bench/). Would someone be able to please point me to a version of these? Thanks so much! Steve
Steve Benz wrote:
Hi~
I've recently converted our monet install from the cimpl version of the php interface to the new native socket interface. I see the readme references some benchmarks at a url that does not appear to exist (http://www.ins.cwi.nl/~gm/bench/). Would someone be able to please point me to a version of these?
Thank you for reporting. Martin
Thanks so much! Steve
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
This URL was never going to be accessible from the outside world. The files are now available at http://homepages.cwi.nl/~gm/monetdb/bench/ . On 2010-03-25 16:00, Martin Kersten wrote:
Steve Benz wrote:
Hi~
I've recently converted our monet install from the cimpl version of the php interface to the new native socket interface. I see the readme references some benchmarks at a url that does not appear to exist (http://www.ins.cwi.nl/~gm/bench/). Would someone be able to please point me to a version of these?
Thank you for reporting. Martin
Thanks so much! Steve
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
-- Sjoerd Mullender
Great, thanks! So I take this to mean the native socket interface is actually slower? What was the motivation behind switching the main php interface to the socket-based one? On Mar 25, 2010, at 9:41 AM, Sjoerd Mullender wrote:
This URL was never going to be accessible from the outside world. The files are now available at http://homepages.cwi.nl/~gm/monetdb/bench/ .
On 2010-03-25 16:00, Martin Kersten wrote:
Steve Benz wrote:
Hi~
I've recently converted our monet install from the cimpl version of the php interface to the new native socket interface. I see the readme references some benchmarks at a url that does not appear to exist (http://www.ins.cwi.nl/~gm/bench/). Would someone be able to please point me to a version of these?
Thank you for reporting. Martin
Thanks so much! Steve
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev _______________________________________________ MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
-- Sjoerd Mullender
------------------------------------------------------------------------------ Download Intel® Parallel Studio Eval Try the new software tools for yourself. Speed compiling, find bugs proactively, and fine-tune applications for parallel performance. See why Intel Parallel Studio got high marks during beta. http://p.sf.net/sfu/intel-sw-dev____________________________________________... MonetDB-users mailing list MonetDB-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/monetdb-users
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Op 25-03-10 18:03, Steve Benz schreef:
So I take this to mean the native socket interface is actually slower? What was the motivation behind switching the main php interface to the socket-based one?
You wouldn't be required to have the Mapi client library on your system to communicate to a MonetDB database. Although it is interesting that it is slower; didn't look at the Pecl code yet of this extension, but in general these kind of solutions don't have to be slower. Stefan -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.14 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEAREKAAYFAkurnKQACgkQYH1+F2Rqwn2x1wCfcP4tpDMTYgLBz5PuhMzvqgqG OS8AnAvcTxdji6qfkeiCLu0E2LAJhLNT =zKtY -----END PGP SIGNATURE-----
On Thu, Mar 25, 2010 at 1:25 PM, Stefan de Konink
Although it is interesting that it is slower; didn't look at the Pecl code yet of this extension, but in general these kind of solutions don't have to be slower.
For the python client I found setting socket options made a huge speed difference. m
On Thu, Mar 25, 2010 at 1:43 PM, Steve Benz
Interesting... mind sharing some of those options you messed with? Were they on the server side or the client side?
Client. See this ticket: http://sourceforge.net/tracker/index.php?func=detail&aid=2925750&group_id=56967&atid=482468 m
participants (5)
-
Mark Bucciarelli -
Martin Kersten -
Sjoerd Mullender -
Stefan de Konink -
Steve Benz