| To: | Larry McVoy <lm@xxxxxxxxxxxx> |
|---|---|
| Subject: | Re: TOE brain dump |
| From: | Timothy Miller <miller@xxxxxxxxxxxxxx> |
| Date: | Tue, 05 Aug 2003 15:28:09 -0400 |
| Cc: | David Lang <david.lang@xxxxxxxxxxxxxxxxxx>, Erik Andersen <andersen@xxxxxxxxxxxx>, Werner Almesberger <werner@xxxxxxxxxxxxxxx>, Jeff Garzik <jgarzik@xxxxxxxxx>, netdev@xxxxxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, Nivedita Singhvi <niv@xxxxxxxxxx> |
| References: | <20030803194011.GA8324@work.bitmover.com> <Pine.LNX.4.44.0308031253240.24695-100000@dlang.diginsite.com> <20030803203051.GA9057@work.bitmover.com> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.0.1) Gecko/20020823 Netscape/7.0 |
|
Larry McVoy wrote: On Sun, Aug 03, 2003 at 01:13:24PM -0700, David Lang wrote: Physical space? Power usage? Heat dissipation? Optimization for the specific task? Fast, low latency communication between CPU and device (ie. local bus)? Maintenance? Lots of reasons why one might pay more for the offload card. If you're cheap, you'll just use the software stack and a $10 NIC and just live with the corresponding CPU usage. If you're a performance freak, you'll spend whatever you have to to squeeze out every last bit of performance you can. Mind you, another option is, if you're dealing with the kind of load that requires that much network performance, is to use redundant servers, like google. No one server is exceptionally fast, but it not many people are using it, it's fast enough. |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: TOE brain dump, Ingo Oeser |
|---|---|
| Next by Date: | [PATCH] (2/2) Convert ROSE to seq_file, Stephen Hemminger |
| Previous by Thread: | Re: TOE brain dump, Jeff Garzik |
| Next by Thread: | Re: TOE brain dump, jamal |
| Indexes: | [Date] [Thread] [Top] [All Lists] |