[Beowulf] Benchmark between Dell Poweredge 1950 And 1435

Robert G. Brown rgb at phy.duke.edu
Mon Mar 12 12:16:43 PDT 2007


On Mon, 12 Mar 2007, Greg Lindahl wrote:

> On Mon, Mar 12, 2007 at 01:40:27PM -0400, Robert G. Brown wrote:
>
>> Have you looked over the HT consortium site?  It is truly a lovely spec
>> but they didn't quite make the most of their brief window of opportunity
>> to kill off the bus.  Everybody went PCI-X instead of HT.
>
> Guys, this is misunderstanding history. AMD was not positioning HT as
> an expansion bus until we came along. Even after that, it was clear
> that PCI Express was going to have most of the market, so it's only the
> folks with high-end needs who might even consider directly attaching to HT.
>
> No FUD required, no lack of energy, nothing weird, and this isn't a
> *bad* result. It just is. The highest volume solution tends to be
> the one that's good enough, and PCI Express is good enough for most
> uses.

I don't disagree, although even their earliest pictures had more than
just CPU and memory hanging off the network.  As in CPU, memory, a PCI
bus and several direct attached devices were on most of the
transparencies that I recall of AMD's roadmap presentation 2-3 years
ago.  I think that even then, they had a gleam in their eye, although
they also may have been realistic about their chances...

You might want to edit the widipedia page on this, as it states
"Hypertransport has largely fallen out of favor with the networking
community, in favor of SPI 4.2 and PCI-Express" which definitely implies
that at one time it was IN favor...:-)

     rgb

>
> -- greg
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
>

-- 
Robert G. Brown	                       http://www.phy.duke.edu/~rgb/
Duke University Dept. of Physics, Box 90305
Durham, N.C. 27708-0305
Phone: 1-919-660-2567  Fax: 919-660-2525     email:rgb at phy.duke.edu





More information about the Beowulf mailing list