lkml.org 
[lkml]   [1996]   [Nov]   [20]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
From
SubjectRe: FYI: I2O Architecture (fwd)
Date

Followup to: <Pine.LNX.3.91.961119200905.74A-100000@nextd.demon.co.uk>
By author: Mark Hemment <markhe@nextd.demon.co.uk>
In newsgroup: linux.dev.kernel
>
> With the processing power that will be on some of the I2O cards, we
> might find network protocol stacks (TCP?), etc, being off-loaded
> from the central CPUs. Would this be a step backwards or forwards
> from SMP?
>

Backwards. This is a classic fallacy, described in detail in most
computer architecture courses. Basically, the problem is that main
CPU's tend to improve faster than auxilliary I/O processors, and that
communication schemes between the two tend to be too limiting for
future growth.

-hpa
--
This space intentionally has nothing but text explaining why this
space has nothing but text explaining why this space would otherwise
have been left blank, and would otherwise have been left blank.


\
 
 \ /
  Last update: 2005-03-22 13:38    [W:0.187 / U:0.692 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site