[extropy-chat] IBM's Blue Gene

[Eugen Leitl]

On Sat, Apr 16, 2005 at 10:19:24PM -0400, Rik van Riel wrote:

> I'm not convinced that the Cell will matter all that much,
> the bottleneck for computations is shifting away from the
> CPU core and onto memory bandwidth.  The Cell CPU doesn't

Yes, most number crunch applications are memory-starved.

> help memory bandwidth one bit...

Actually it does, because each core is fed by its own on-die memory.
Now if Cell did come with a fully switched 6-link interconnect...

-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a>
______________________________________________________________
ICBM: 48.07078, 11.61144            http://www.leitl.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE
http://moleculardevices.org         http://nanomachines.net
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: not available
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20050417/89052167/attachment.bin>