By: aaron spink (aaronspink.delete@this.notearthlink.net), July 27, 2012 9:36 am
Room: Moderated Discussions
jp (jipe4153.delete@this.gmail.com) on July 27, 2012 9:18 am wrote:
> About bandwidth, the GPUs already
> have the fastest RAM out there ( over 250 GB/s ) and they have no reason not to
> continue this lead (read mentioned FLOPS/bandwidth ratio "issue")
>
GPUs have high local bandwidth but rather poor global bandwidth. Not to mention rather limited capacity.
> Looking at the articles at hpcwire about
> new clusters over the last 1.5 years its obvious that almost everyone is buying
> Nvidia Tesla cards. In fact ORNL:s newest cluster (fastest in the US) will be
> based on the new Kepler cards (K20).
>
Very few are buying Tesla cards. And most of the data says they are not any better than CPUs with much more complex programming models.
> About bandwidth, the GPUs already
> have the fastest RAM out there ( over 250 GB/s ) and they have no reason not to
> continue this lead (read mentioned FLOPS/bandwidth ratio "issue")
>
GPUs have high local bandwidth but rather poor global bandwidth. Not to mention rather limited capacity.
> Looking at the articles at hpcwire about
> new clusters over the last 1.5 years its obvious that almost everyone is buying
> Nvidia Tesla cards. In fact ORNL:s newest cluster (fastest in the US) will be
> based on the new Kepler cards (K20).
>
Very few are buying Tesla cards. And most of the data says they are not any better than CPUs with much more complex programming models.
