lkml.org 
[lkml]   [1998]   [Oct]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Help clear up a misconception about SMP systems
On Sat, 10 Oct 1998, Mike A. Harris wrote:

> For one, any programs you have, unless multithreaded will sit on
> one CPU or the other. So if you boot up, and run program "Foo",
> unless foo is multithreaded, it will run the same as on UP
> system, perhaps even a bit slower. If you have a bunch of
> processes running, however, each individual process will run no
> faster than on a single CPU system, but you can run more
> processes now. Multithreaded applications can take full power of
> the multi CPU though, since different threads can operating on
> different CPU's.

Program != process. Program can be multithreaded or can consist of
multiple processes that may talk to each other. As I understand it,
interprocess communication in SMP causes less context switches than on UP,
therefore the programs that consist of multiple processes should benefit
from it. OTOH, imperfect kernel locks granularity causes delays while
process/thread is in the kernel, and all kinds of programs regardless of
their processes/threads design suffer from that.

--
Alex


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:44    [W:0.146 / U:0.036 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site