lkml.org 
[lkml]   [1999]   [Jul]   [31]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: Scheduling latencies news: less RAM = less latency
On Sun, Aug 01, 1999 at 03:01:58AM +0000, Steve Underwood wrote:
> Yes, such cache thrashing is certainly an issue. The trick is to strike the right
> balance, rather than avoid improving latencies. The decision you really want is not
> "is there someone ready to run" but "is there someone ready to run, that is about
> to kick off some slow I/O that could be left running while I continue this work".
> Unfortunately that's not so easy to achieve.
>

And there are many people who believe that a couple of percent slowdown
in the normal case to make this work better is worth it. Unfortunately,
"this" changes and the percents add up.

Some of this is pure cost shifting. I get this all the time for the RT
kernel -- "you must add feature X so we don't have to rewrite our code
that runs like the tranquilized fossil of a giant ground sloth
on Chorus/Solaris/... ." It's hard to strike the correct balance between
making the OS useful and subsidizing other peoples bad legacy software.



-
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 13:53    [W:0.036 / U:3.192 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site