Re: Illusory Multicore Support in 5.2?
- To: mathgroup at smc.vnet.net
- Subject: [mg66128] Re: Illusory Multicore Support in 5.2?
- From: "Chris Chiasson" <chris.chiasson at gmail.com>
- Date: Sun, 30 Apr 2006 04:21:48 -0400 (EDT)
- References: <200604260838.EAA02742@smc.vnet.net><firstname.lastname@example.org> <200604270836.EAA01610@smc.vnet.net> <email@example.com> <firstname.lastname@example.org>
- Sender: owner-wri-mathgroup at wolfram.com
I could guess. Mathematica likes to eat up 100% cpu and hog all of the memory for a variety of kernel inputs (like some problems where it is inverting a large symbolic matrix, or whatever). This is not to say that I think the problem should be solvable, but maybe that the computer should at least remain responsive. Maybe they decided to set # kernel threads = # processors - 1 to improve responsiveness under high load? Like I said, a guess - from a non-expert.