On 04/30/2014 10:02 AM, Prentice Bisbal wrote:
On 04/30/2014 02:25 AM, Bill Broadley wrote:
* will 3rd parties like nvidia be able to do something compelling with
cache coherent access to main memory?
This seems very likely, as NVidia is part of the founding members of
the OpenPOWER consortium and IBM and NVidia announced plans to
integrate POWER processors and Tesla GPUs at SC13:
http://nvidianews.nvidia.com/Releases/IBM-NVIDIA-to-Supercharge-Corporate-Data-Center-Applications-and-Next-Generation-Supercomputers-a65.aspx
http://www-03.ibm.com/press/us/en/pressrelease/41684.wss
Since the POWER8 processor is part of the OpenPOWER initiative, it is
my understanding that any vendor should be able to license the POWER8
IP and extend it, similar to the ARM processor.
More on this topic from HPCwire:
"BM is making a big play in hybrid computing, seeking to marry its
POWER8 processors with various kinds of accelerators and high-speed
networking and opening up its chip and system software through the
OpenPOWER Foundation. "
and
"IBM wants for all manner of accelerators to link very tightly with
POWER8 processors and for scale-out clusters based on these chips to
have the right kind of acceleration – be it a GPU, an FPGA, a DSP, or
some other component – necessary for their particular jobs."
http://www.hpcwire.com/2014/04/23/power8-openpower-might-mean-hpc/
_______________________________________________
Beowulf mailing list, [email protected] sponsored by Penguin Computing
To change your subscription (digest mode or unsubscribe) visit
http://www.beowulf.org/mailman/listinfo/beowulf
