[Beowulf] Roadrunner shutdown
prentice.bisbal at rutgers.edu
Thu Apr 4 11:04:23 PDT 2013
IBM pulled out of the proposal, not NCSA, and it was because it wasn't
profitable enough for IBM to continue, not because IBM's solution was
too expensive, or wasn't working. I wouldn't call the Cray solution
commodity, either. It uses a proprietary network, and Keplers used it in
don't look like normal, off-the-shelf PCIE packaging in the videos I've
seen. I could wrong about that last part, though. I'm sure the
processors themselves are the same - I'm just talking about the
packaging of the PCB.
I'm CC'ing this reply back to the list because I think this is a good
discussion for the whole list, and I don't see any reason for you to
reply to me off-list on this topic. Our conversation has been
professional, more or less, so far. Sorry if you disagree.
On 04/04/2013 01:12 PM, Bret Stouder wrote:
> When you get a moment reach out to the fine folks at NCSA and get there prospective of why they built Blue Waters on commodity products from Cray instead of the IBM proposal. Again, my position is pretty simple......you can get a heck of a lot more computational cycles for the money by staying in a commodity product as opposed to buying proprietary product from the Big Blue. If you have PHI questions I am more than happy to take those offline with you as we do have experience with PHI, we have deployed it and we can offer you feedback. Not at all offering any solicitation here, just making sure you are aware that we do have data that we can share if you are interested.
> From: beowulf-bounces at beowulf.org [beowulf-bounces at beowulf.org] On Behalf Of Prentice Bisbal [prentice.bisbal at rutgers.edu]
> Sent: Thursday, April 04, 2013 8:22 AM
> To: beowulf at beowulf.org
> Subject: Re: [Beowulf] Roadrunner shutdown
> On 04/03/2013 11:42 AM, Geoffrey Jacobs wrote:
>> On 04/03/2013 08:27 AM, Bret Stouder wrote:
>>> For the money you could buy a lot more real computational cycles that last a lot longer than anything proprietary from IBM. The reason the Origins have been out of production for a long time is the result of research moving to a commodity server platform. Taken a look at what Cray offers lately? Looks a lot like commodity. If you are going to take the time to port code, it seems that a reasonable argument could be made to stay away from Blue Gene and port to Cuda or Phi.
> Actually, that's a naive argument. The Blue Gene architecture uses
> standard MPI and support C, C++, Fortran, and Python, so any MPI
> compliant program written in any of those languages, which will also run
> just fine on a Blue Gene, with no porting necessary - just compile the
> code for the BG processors using GCC or the IBM XL compilers. IBM
> provides the ESSL and MASSV libaries that are optimized for the Blue
> Gene, but those are specified at link time, so I wouldn't really call
> that porting. Sure, there are tweaks you could use to really optimize
> your code for BG, but just doing the above will get you very far.
> In contrast, Cuda requires adding Cuda specific code to move data into
> and out of the device, and the sections to be run on the cuda processors
> needs to be rewritten in terms of cuda thread index, etc.
> For a large code, this could take a significant amount of labor, enough
> that it's conceivable in certain cases buying a Blue Gene could be
> quicker and more cost-effective than spending all the money on the labor
> to port your code to Cuda, test, optimize, test...
> In short, it's a lot more effort to port MPI-compliant code for Cuda
> than for a Blue Gene.
> I specifically didn't mention Phi because I do not know the details of
> porting/optimizing code for Phi.
> Beowulf mailing list, Beowulf at beowulf.org sponsored by Penguin Computing
> To change your subscription (digest mode or unsubscribe) visit http://www.beowulf.org/mailman/listinfo/beowulf
More information about the Beowulf