Template:CheahaTflops

From UABgrid Documentation
(Difference between revisions)
Jump to: navigation, search
(correct flops/cycle for sandy bridge (gen4 and gen5))
(added TFLOP explanation)
 
(15 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 +
[http://en.wikipedia.org/wiki/FLOPS#Computing Theoretical Peak Flops] = (number of cores) * (clock speed) * (instructions per cycle)
 +
 
{| class="wikitable"
 
{| class="wikitable"
|-
 
!Generation!! Type!! Nodes!! CPUs per Node!! Cores Per CPU!! Total Cores!!Clock Speed (GHz)!! Instructions Per Cycle !!Hardware Reference
 
 
|-
 
|-
|Gen 2|| Intel Xeon E5450||align="right"| 24||align="right"| 2||align="right"| 4||align="right"| 192||align="right"| 3.00||align="right"| 4||[http://ark.intel.com/products/33083/Intel-Xeon-Processor-E5450-(12M-Cache-3_00-GHz-1333-MHz-FSB) Intel Xeon E5450]
+
!Generation!! Cpu Tflops Per Node!! Gpu Tflops Per Node!! Tflops Per Node!! Nodes!! Tflops
 
|-
 
|-
|Gen 3|| Intel Xeon X5650||align="right"| 48||align="right"| 2||align="right"| 6||align="right"| 576||align="right"| 2.66||align="right"| 4|| [http://ark.intel.com/products/47922/Intel-Xeon-Processor-X5650-(12M-Cache-2_66-GHz-6_40-GTs-Intel-QPI) Intel Xeon E6550]
+
|6||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 14||align="right"| 13.44
 
|-
 
|-
|Gen 4|| Intel Xeon E5-2680||align="right"| 3||align="right"| 2||align="right"| 8||align="right"| 48||align="right"| 2.70||align="right"| 8|| [http://ark.intel.com/products/64583 Intel Xeon E2680]
+
|6||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 38||align="right"| 36.48
|
+
 
|-
 
|-
|Gen 5|| Intel Xeon E5-2650||align="right"| 12||align="right"| 2||align="right"| 8||align="right"| 192||align="right"| 2.00||align="right"| 8|| [http://ark.intel.com/products/64590/Intel-Xeon-Processor-E5-2650-20M-Cache-2_00-GHz-8_00-GTs-Intel-QPI Intel Xeon E2650] (Nodes dedicated to [https://dev.uabgrid.uab.edu/wiki/OpenStackPlusCeph OpenStack+Ceph] with 10Gbs network)
+
|6||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 44||align="right"| 42.24
|}
+
|-
 
+
|7||align="right"| 1.08||align="right"| 17.06||align="right"| 18.14||align="right"| 18||align="right"| 326.43
{| class="wikitable"
+
|-
|+Theoretical Peak Flops = (number of cores) * (clock speed) * (instructions per cycle)
+
|8||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 21||align="right"| 20.16
 
|-
 
|-
!Generation!!Theoretical Peak Tera-FLOPS
+
|8||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 10||align="right"| 9.60
 
|-
 
|-
|Gen 2||align="right"| 2.304
+
|8||align="right"| 0.96||align="right"| ||align="right"| 0.96||align="right"| 4||align="right"| 3.84
 
|-
 
|-
|Gen 3||align="right"| 6.129
+
|9||align="right"| 2.30||align="right"| ||align="right"| 2.30||align="right"| 52||align="right"| 119.81
 
|-
 
|-
|Gen 4||align="right"| 1.036
+
|10||align="right"| 4.10||align="right"| ||align="right"| 4.10||align="right"| 34||align="right"| 139.26
 
|-
 
|-
|Gen 5||align="right"| 3.072
+
|TOTAL||align="right"| ||align="right"| ||align="right"| ||align="right"| ||align="right"| 711.26
 
|-
 
|-
|'''Total TFLOPS'''||align="right"| 12.541
 
 
|}
 
|}

Latest revision as of 12:24, 18 August 2021

Theoretical Peak Flops = (number of cores) * (clock speed) * (instructions per cycle)

Generation Cpu Tflops Per Node Gpu Tflops Per Node Tflops Per Node Nodes Tflops
6 0.96 0.96 14 13.44
6 0.96 0.96 38 36.48
6 0.96 0.96 44 42.24
7 1.08 17.06 18.14 18 326.43
8 0.96 0.96 21 20.16
8 0.96 0.96 10 9.60
8 0.96 0.96 4 3.84
9 2.30 2.30 52 119.81
10 4.10 4.10 34 139.26
TOTAL 711.26
Personal tools
Namespaces

Variants
Actions
Navigation
Toolbox