This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision Next revision Both sides next revision | ||
cluster:112 [2013/04/12 19:32] hmeij [Dell Jan-2013] |
cluster:112 [2013/05/13 14:35] hmeij [Dell April-2013] |
||
---|---|---|---|
Line 19: | Line 19: | ||
| Cooling (Btus((per hour, 1,000 Btus=293 Watts per hour)))| | | Cooling (Btus((per hour, 1,000 Btus=293 Watts per hour)))| | ||
| Annual Costs((power and cooling)) ($)| | 46, | | Annual Costs((power and cooling)) ($)| | 46, | ||
+ | | $1, | ||
Line 60: | Line 61: | ||
* " | * " | ||
- | ===== Dell Jan-2013 ===== | ||
- | ** Actual data obtained 11 April - 18 April, 2013 ** | ||
- | |||
- | Watt-o-Meter Stats; pulled one power supply ... mean/1000 * 24 * 365 * 30! | ||
- | |||
- | ^ Node^ threads^ | ||
- | | c04| 8| 2.29| 06: | ||
- | | c32| 8| 6.94| 19: | ||
- | | c27| 8| 7.42| 19: | ||
- | | c10| 8| 10.21| | ||
Line 99: | Line 90: | ||
** This is what the picture would look like using the 2009 spreadsheet data** | ** This is what the picture would look like using the 2009 spreadsheet data** | ||
- | * 240 cores, make them all ethernet based | + | What we should look at is teraflops |
- | * slight change but doable | + | |
- | * 1U yields 16 cores, 15U yields 240 cores with a total of 1,920 GB memory | + | |
- | * major change, Dell offers 388 GB | + | |
- | * acquisition costs | + | |
- | * 15 x $5,782 = $86,730 + PDUs, ethernet switch, cables | + | |
- | * power/AC costs: $23, | + | |
- | * add both costs of new hardware and divide by dell total annual costs power+cooling | + | |
- | * ROI: 3.4 years - but this is on a core by core basis | + | |
- | + | ||
- | What we should look at is teraflops... | + | |
- | * 0.66 teraflops, measured with Linpack | + | * 0.66 teraflops, measured with Linpack |
- | * E5-2660 is rated at 140 Gflops so one node provides 280 Gflops (theoretically) | + | * E5-2660 is rated at 140 Gflops so one node provides 280 Gflops (theoretical) |
* 2 and 1/3rd 1U nodes then replace the cluster | * 2 and 1/3rd 1U nodes then replace the cluster | ||
* acquisition costs | * acquisition costs | ||
Line 121: | Line 102: | ||
* ROI: 3/4 year - but this is on a teraflop by teraflop basis | * ROI: 3/4 year - but this is on a teraflop by teraflop basis | ||
- | And if we double the teraflops...1.6 Tflops (actually +1Tflops), 1.5 years ROI for 96 cores. | + | And if we double the teraflops...1.6 Tflops (actually + 1 Tflops), 1.5 years ROI for 96 cores. |
Line 137: | Line 118: | ||
* LCBR 06 | * LCBR 06 | ||
* (#3 on Enterprise UPS) PDI1 CIR 31,33 | * (#3 on Enterprise UPS) PDI1 CIR 31,33 | ||
+ | |||
+ | |||
+ | ====== Dell April-2013 ====== | ||
+ | |||
+ | ** Actual data obtained April - May, 2013 ** | ||
+ | |||
+ | Kill-A-Watt Meter Stats; pulled one power supply ... mean/1000 * 24 * 365 * 30! | ||
+ | |||
+ | ^ Node^ jobs^ kwh^ time hh: | ||
+ | | c04| 8| 2.29| 06: | ||
+ | | c32| 8| 6.94| 19: | ||
+ | | c27| 8| 7.42| 19: | ||
+ | | c06| 8| 27.46| | ||
+ | | c00| 8| 8.26| 21: | ||
+ | | c03| 8| 10.36| | ||
+ | | c04| 8| 12.02| | ||
+ | | c09| 8| 10.36| | ||
+ | | c10| 8| 10.21| | ||
+ | | c11| 4| | | | still running| | ||
+ | | c13| 3| | | | still running| | ||
+ | | c17| 8| 27.07| | ||
+ | | c18| 8| 11.23| | ||
+ | | c20| 8| 50.23| | ||
+ | | c23| 8| 8.71| 24: | ||
+ | | c25| 8| 9.80| 25: | ||
+ | | c29| 8| 30.22| | ||
+ | | c31| 8| 10.30| | ||
+ | | c32| 8| 11.42| | ||
+ | | c33| 8| 46.23| | ||
+ | | c35| 8| 13.69| | ||
+ | |||
+ | |||
+ | |||
+ | |||
+ | |||
\\ | \\ | ||
**[[cluster: | **[[cluster: |