[Beowulf] SiCortex experience anyone?

[Gerry Creager]

Steve Cousins wrote:
> 
> 
> On Tue, 27 May 2008, Mark Hahn wrote:
> 
>>>> Does anyone have any experience with SiCortex machines? Any 
>>>> thoughts? They
>>>> look cool and they don't use much power but I wonder how they 
>>>> compare to
>>>> blade type systems.
>>>
>> eh?  blade systems are just tweaks for packaging and cable-management.
>> I don't believe they have much if any advantage over well-designed
>> non-blade hardware.  or did you mean that both blades and sicortex
>> are basically boutique approaches, as opposed to fungible/commodity?
> 
> 
> I was using the blade example just because of the similarity of having a 
> stand-alone system with a single power source.

Different paradigms save in the idea that Blades are simply a way to 
concentrate heat and make you always come back to one vendor for 
additions or replacements.  I don't like to lock into a single vendor 
unless they're well proven.  I've made the blade mistake with a vendor 
who, overall, always did the right thing, but having a bunch of blades 
did little more than look cool when I needed to expand.

gerry

>>>> I know they are not Beowulfs but being distributed
>>>> memory machines I figured someone here might have real-world 
>>>> experience with
>>>> one.
>>>
>> I think it's still a beowulf if you apply "the beowulf approach"
>> (linux and open-source HPC clustering) to non-commodity hardware.
>> we still have a beowulf of Compaq ES40 alphas, for instance.
> 
> 
> Great. I'm glad my post was on topic then. My view is that the SiCortex 
> machine is not a Beowulf because the term "node" seems to be a chip on 
> the same board as all of the other nodes. You can't really add nodes 
> like you could add an ES40 to your Compaq Beowulf.  Of course opinions 
> will vary. I suppose you could add other SiCortex machines... Can you?
> 
> Do you have any experience with them? How easy is it to port code to it? 
> Performance characteristics with certain types of code?
> 
> Thanks,
> 
> Steve
> _______________________________________________
> Beowulf mailing list, Beowulf at beowulf.org
> To change your subscription (digest mode or unsubscribe) visit 
> http://www.beowulf.org/mailman/listinfo/beowulf

-- 
Gerry Creager -- gerry.creager at tamu.edu
Texas Mesonet -- AATLT, Texas A&M University
Cell: 979.229.5301 Office: 979.862.3982 FAX: 979.862.3983
Office: 1700 Research Parkway Ste 160, TAMU, College Station, TX 77843