Differences

This shows you the differences between two versions of the page.

--- cluster:167 [2018/08/01 12:40]
hmeij07 [CPU vs GPU]
+++ cluster:167 [2018/08/01 14:11] (current)
hmeij07 [July 2018]
@@ Line 59: / Line 59: @@
 We shall wait for Jul18 metrics.
-==== Jul 2018 ====
+==== July 2018 ====
 ^ Metric ^ CPU ^ Ratio ^ GPU ^ Notes July 2017 ^
@@ Line 68: / Line 68: @@
 | Job Count | 12,798  |  18:1  |  722 | processed jobs irregardless of exit status |
 | Avail Hours | 898,752  |  50:1  |  17,856 | total cpu cores, total gpus |
-| Job Hours |  260,997  |  69:1  |  3,805 | cumulative hours of consumed usage |
+| Job Hours |  322,207  |  1732:1  |  186 | cumulative hours of consumed usage |
-| Job Hours % | 30  |  1:1  |  26 | as a percentage of available |
+| Job Hours % | 36  |  36:1  |  1 | as a percentage of available |
-| Avail Hours2 | 696,384  |  47:1  |  14,880 | total for cpu cores minus hp12's 256 cores, total for gpus |
-| Job Hours2 % | 37  |  1.5:1  |  26 | more realistic...hp12 rarely used |
-Based on the utilization string '0000', meaning all gpus are idle as polled every 15 mins during July 2018, the GTX gpus were 65% completely idle and the K20s were 61% completely idle.
+Based on the utilization string '0000', meaning all gpus are idle on a single node as polled every 10 mins during July 2018, the GTX gpus were 65% completely idle and the K20s were 61% completely idle.
 **[[cluster:0|Back]]**

DokuWiki