Quoting Joe Landman <[EMAIL PROTECTED]>, on Tue 27 May
2008 05:52:56 PM PDT:
Jim Lux wrote:
The value of a supercomputer under the desk is primarily to the end
user (faster whatever to free up an expensive engineer's time).
They don't see much difference between buying the software for
Jim Lux wrote:
The value of a supercomputer under the desk is primarily to the end user
(faster whatever to free up an expensive engineer's time). They don't
see much difference between buying the software for a single processor
or multiprocessor. The ISV won't get many incremental sales from
Quoting "Lombard, David N" <[EMAIL PROTECTED]>, on Tue 27 May
2008 05:02:30 PM PDT:
I wasn't implying that ISVs were the cause of their demise. I was
implying that they couldn't build a convincing case for them to get on
board.
To put a finer point on this, ISVs ask: How do I get increme
On Tue, May 27, 2008 at 04:52:31PM -0400, Joe Landman wrote:
> Greg Lindahl wrote:
> >On Tue, May 27, 2008 at 04:12:35PM -0400, Joe Landman wrote:
> >
> >Orion had poor price/performance. ISVs were a symptom, not a cause.
>
> I agree they had poor price performance. ISVs always (have to) ask the
I guess I am the token SiCortex person who posts here sometimes.
There are probably other
lurkers as well. Hopefully this is generic enough to not trigger a
commercial
allergic response from the moderator...
The philosophical idea was something like noticing that HPC codes tend
to
miss i
Joe Landman <[EMAIL PROTECTED]> writes:
>> I think you're looking through the lens of your customer base. A large
>
> Quite possibly.
>
>> number of clusters never run commercial codes, and some commercial
>
> True. They tend to be low margin commodity clusters that are very
> hard to support "va
Greg Lindahl wrote:
On Tue, May 27, 2008 at 04:12:35PM -0400, Joe Landman wrote:
For them to be successful, they need customers to buy machines, which
usually means getting commercial codes onto them.
I think you're looking through the lens of your customer base. A large
Quite possibly.
n
On Tue, May 27, 2008 at 04:12:35PM -0400, Joe Landman wrote:
> For them to be successful, they need customers to buy machines, which
> usually means getting commercial codes onto them.
I think you're looking through the lens of your customer base. A large
number of clusters never run commercial
Gerry Creager wrote:
SiCortex is not too far from the Beowulf concept but is possibly closer
to a packaged single DM entity. I'd thought of it as a low-power, low
heat (DM) HPC system without ever considering the beowulfness of it. It
uses Linux and MPI, so it meets my needs in that regard.
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-ma
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
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 believ
The ISO for the ubuntu (server, AMD64) install was 550MB (the other day). So
1GB seems like alot to me.
Of course, as I've mentioned in the past, I first booted System V with 512K
(on a 286). No room for applications, "vi temp" hung (decades later, it
occurs to me it may have thrashed to virtual m
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 wel
14 matches
Mail list logo
