[Beowulf] Opteron

Joe Landman landman at scalableinformatics.com
Fri Nov 16 07:25:56 PST 2007

Mark Hahn wrote:
>> Come on then SC07'ers. Whats the buzz with barcelona?
> I didn't get to go dammit, but I think Intel used Penryn reporting
> extremely effectively.  I'm personally astonished how poorly AMD is 
> doing in keeping up, or at least appearing to keep up.  so far,
> it's quite sad: only very low-clock chips and no real promise of any
> upcoming save-the-bacon feature.  AFAIKT, Intel has pretty much 

Hmmm... my impression was/is severe issues in execution.  The technology 
is there, it seems that some in AMD (not the folks at the show) just 
"don't get it"(TM).  I am guessing they think HPC is a small niche 
market which is not important to them.

This is a shame, as they do have very good technology that, if they can 
just deliver it, would be quite competitive and leading in a number of ways.

> destroyed AMD's lead in memory performance now.  unless Hf gates start 

I am not sure about this.  Have some tests to do.  Bug Doug Eadline on 
this, he has some good early impressions based upon his testing.

> turning to cheddar, or Quickpath fails utterly, I really don't
> see AMD surviving.

A little extreme I think.

> any benefit from the ATI merger is more than a few years off, IMO,

Some of us doubt there ever will be any benefit from this merger.

> and quite specialized.  Fusion may be a useful feature for markets where 
> a highly integrated solution is rewarded (cheap desktops, laptops).
> so far, at least, it's just "chip legos", which is not what I'd call
> a Big Vision.  Intel's reported approach of x86-ISA-based GPUs sounds
> MUCH more exciting, since it could be hopefully be subverted to do HPC
> (and without the sadistic computing model of current GP-GPU.)
> who knows?  maybe AMD will do a quick merger with TSMC, start producing 
> K10's next month with 8M onchip L3 using TSMC's 55nm process,
> run at 3.5 GHz with ddr3/1600.  short of something fairly dramatic in 
> those directions, it's hard to see reasons not to buy Intel.
> I'm more interested in hearing whether 10G eth made a good showing
> vs IB.

I saw the Mellanox 10GbE units, Myricom's stuff,  the Arastra (sp?) 
switches, Woven systems stuff.  For 10GbE to be meaningful, it has to be 
port cost competitive with IB.  Three of the vendors above grasped that. 
  There were a number of vendors that (mistakenly) believe that $2k 
USD/port is a good thing in clusters and storage systems.  I wish them 
luck with that.  Myricom's stuff is pretty neat, well priced.  They have 
some interesting ideas.


Joseph Landman, Ph.D
Founder and CEO
Scalable Informatics LLC,
email: landman at scalableinformatics.com
web  : http://www.scalableinformatics.com
phone: +1 734 786 8423
fax  : +1 866 888 3112
cell : +1 734 612 4615

More information about the Beowulf mailing list