[MonetDB-users] hash accelerator
Hello, I'm working with mal and I've seen in the web documentation, in the part of the inner core that there are built-in accelerator functions. As I can see, there is one search accelerator: hashing. ¿Could you tell me how to use this accelerator using mal language? Looking forward to hearing from you. Thank you. Mike
Hello again, doing different trials I have found this message when I use the option algorithms in trace option:
mserver5 --dbname=bhash3 --algorithms
#BAThash: create hash(4); # MonetDB server v5.18.5, based on kernel v1.36.5 # Serving database 'bhash3', using 8 threads # Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked # Found 31.431 GiB available main-memory. # Copyright (c) 1993-July 2008 CWI. # Copyright (c) August 2008-2010 MonetDB B.V., all rights reserved #BAThash: create hash(24); # Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
#BAT_select_(b=M5system_auth_scen): BATslice(v=M5system_auth_scen, low=0, high=0);
Could it means that I have activated the built-in accelerator functions? Thank you. Mike El jue, 17-06-2010 a las 08:27 +0200, Miguel Ángel Águila Lorente escribió:
Hello,
I'm working with mal and I've seen in the web documentation, in the part of the inner core that there are built-in accelerator functions. As I can see, there is one search accelerator: hashing. ¿Could you tell me how to use this accelerator using mal language?
Looking forward to hearing from you. Thank you.
Mike
Hi Mike, MonetDB decides internally automatically whether or not to build hash indices on the fly. Usually this happens in operation like join, duplicate elimination, grouping, or a like that need to match multiple against multiple values. Indeed that's what the --algorithm log shows. (Note though, that the --algorithm option is mainly for debugging purpose. Its output is not complete, i.e., only selected operations log their internal decisions when --algorithm is activated, and the output is hardly useful unless you are familiar with the MonetDB code.) There are no option to activate hash indices explicitely. Definitely not in SQL, and most probably also not in MAL. My questions are, (1) why (or for what purpose) are you using MAL (rather than SQL), and (2) why do you have/feel the need to create hash indices explicitely, i.e., do you (and if so in which cases) experience performance "problems" with MonetDB? Stefan On Fri, Jul 09, 2010 at 11:35:20AM +0200, Miguel Ángel Águila Lorente wrote:
Hello again,
doing different trials I have found this message when I use the option algorithms in trace option:
mserver5 --dbname=bhash3 --algorithms
#BAThash: create hash(4); # MonetDB server v5.18.5, based on kernel v1.36.5 # Serving database 'bhash3', using 8 threads # Compiled for x86_64-unknown-linux-gnu/64bit with 64bit OIDs dynamically linked # Found 31.431 GiB available main-memory. # Copyright (c) 1993-July 2008 CWI. # Copyright (c) August 2008-2010 MonetDB B.V., all rights reserved #BAThash: create hash(24); # Listening for connection requests on mapi:monetdb://127.0.0.1:50000/
#BAT_select_(b=M5system_auth_scen): BATslice(v=M5system_auth_scen, low=0, high=0);
Could it means that I have activated the built-in accelerator functions?
Thank you.
Mike
El jue, 17-06-2010 a las 08:27 +0200, Miguel Ángel Águila Lorente escribió:
Hello,
I'm working with mal and I've seen in the web documentation, in the part of the inner core that there are built-in accelerator functions. As I can see, there is one search accelerator: hashing. ¿Could you tell me how to use this accelerator using mal language?
Looking forward to hearing from you. Thank you.
Mike
------------------------------------------------------------------------------ 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
-- | Dr. Stefan Manegold | mailto:Stefan.Manegold@cwi.nl | | CWI, P.O.Box 94079 | http://www.cwi.nl/~manegold/ | | 1090 GB Amsterdam | Tel.: +31 (20) 592-4212 | | The Netherlands | Fax : +31 (20) 592-4199 |
participants (2)
-
Miguel Ángel Águila Lorente
-
Stefan Manegold