12 Feb
2013
12 Feb
'13
7:14 p.m.
On Tue, Feb 12, 2013 at 7:09 PM, Karl Rupp <[email protected]> wrote:
Which crossover are you referring to? The CPU versus GTX285 at about 20k
dofs, but with only very small gains for another order of magnitude?
I assume it's the cross-over of Xeon Phi vs. Xeon,
MIC is slower than Xeon by more than an order of magnitude at 10k dofs.
but almost all cross-overs happen in the 10k-100k region and are due to PCI-Express latency.
Why is PCI-Express latency important here? Can't the MIC code run entirely on the device?
Scaling this to large clusters, replace Xeon Phi by compute node, PCI-Express by GB-Ethernet/Infiniband as well as change the timescale a bit and you're probably not too far off...