Template:CheahaTflops: Difference between revisions

From Cheaha
Jump to navigation Jump to search
No edit summary
No edit summary
Line 37: Line 37:


<sup>†</sup> Includes four Intel Xeon Phi 7210 accelerators and four NVIDIA K80 GPUs.<br/>
<sup>†</sup> Includes four Intel Xeon Phi 7210 accelerators and four NVIDIA K80 GPUs.<br/>
<sup>††</sup> Includes 72 NVIDIA Tesla P100 16GB GPUs.
<sup>††</sup> Includes 72 [https://images.nvidia.com/content/tesla/pdf/nvidia-tesla-p100-PCIe-datasheet.pdf NVIDIA Tesla P100 16GB] GPUs.

Revision as of 22:02, 15 October 2017

Generation Type Nodes CPUs per Node Cores Per CPU Total Cores Clock Speed (GHz) Instructions Per Cycle Hardware Reference
Gen 2 Intel Xeon E5450 24 2 4 192 3.00 4 Intel Xeon E5450
Gen 3 Intel Xeon X5650 48 2 6 576 2.66 4 Intel Xeon E6550
Gen 4 Intel Xeon E5-2680 3 2 8 48 2.70 8 Intel Xeon E2680
Gen 5 Intel Xeon E5-2650 12 2 8 192 2.00 8 Intel Xeon E2650 (Nodes dedicated to OpenStack+Ceph with 10Gbs network)
Gen 6 Intel Xeon E5-2680 v3 96 2 12 2304 2.50 16 Intel Xeon E5-2680 v3
Gen 7†† Intel Xeon E5-2680 v4 18 2 14 504 2.40 16 Intel Xeon E5-2680 v4
Theoretical Peak Flops = (number of cores) * (clock speed) * (instructions per cycle)
Generation Theoretical Peak Tera-FLOPS
Gen 2 2.304
Gen 3 6.129
Gen 4 1.036
Gen 5 3.072
Gen 6 110
Gen 7†† 358

Includes four Intel Xeon Phi 7210 accelerators and four NVIDIA K80 GPUs.
†† Includes 72 NVIDIA Tesla P100 16GB GPUs.