User Tools

Site Tools


cluster:65

Differences

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

Link to this comparison view

cluster:65 [2008/06/03 13:27] (current)
Line 1: Line 1:
 +\\
 +**[[cluster:​0|Back]]**
 +
 +===== Milestone! =====
 +
 +June 3rd 2008, shortly before 1 pm, the job with JOBID 100,000 was completed. ​
 +
 +^JOBID^USER^STAT^QUEUE^FROM_HOST^EXEC_HOST^JOB_NAME^SUBMIT_TIME^
 +|100000|ztan|DONE|imw|swallowtail|compute-1-23|ising-spring-2008-swt/​simpleRuns/​data018/​t0.900000/​mu-2.630000|Jun ​ 3 12:48|
 +
 +===== Cluster Usage =====
 +
 +Below are some graphs covering the time period 7/​27/​07,​17:​15 to 05/​22/​08,​15:​00 sampled every 5 min intervals.
 +
 +  * Pending Jobs
 +
 +{{:​cluster:​pending.jpg|Pending Jobs}}
 +
 +  * Running Jobs
 +
 +{{:​cluster:​running.jpg|Running Jobs}}
 +
 +The maximum running jobs is 288 (total cores/job slots). ​ The peaks beyond that threshold represents our experimentation with overloading the job slots.
 +
 +===== Fourier =====
 +
 +Francis report ... mathematically,​ a Fourier transform is:
 +
 +F(w) = Integral ( f(t) exp(i*w*t) dt)
 +
 +That is, you integrate a function (or data) with a complex exponential,​ which is just a sum of sine and cosine functions.
 +
 +Practically speaking, suppose f(t) was something like position as a function of time.  Fourier transforming will reveal if there are characteristic frequencies of the data.  For example, if the f(t) was the position of the pendulum in a clock over time, the Fourier transform would show a huge spike at the frequency of the motion, 1 Hz.
 +
 +So Tom is wondering if there are some characteristic frequencies hiding in this usage pattern ... here is the result:
 +
 +{{:​cluster:​fourier.jpg|Fourrier}}
 +
 +===== Mean =====
 +
 +This is the MEAN number of jobs running/​pending over each day.  Averaging over the day eliminates some of the extreme peaks, and gives a good idea of daily usage, ... here is the result:
 +
 +{{:​cluster:​mean.jpg|Mean}}
 +
 +
 +
 +===== Other Stats =====
 +
 +These graphs reflect data that is polled daily. ​ Each day, the stats are calculated that reflect the activity during the last 7 days, for all users, and refer only to jobs whose status is DONE. The last 7 days of data is thus a shifting "​weekly"​ measure; ie not weekly data.
 +
 +Below is the graph for the total number of jobs processed. ​ Below it is a measure called throughput; it reflects the jobs/hour processed. ​ The final graph is the total cpu consumed in seconds.
 +
 +{{:​cluster:​tdjobs.jpg|:​total jobs}}
 +
 +
 +{{:​cluster:​throughput.jpg|:​throughput}}
 +
 +{{:​cluster:​tcpu.jpg|:​total cpu}}
 +
 +
 +
 +\\
 +**[[cluster:​0|Back]]**
 +
  
cluster/65.txt ยท Last modified: 2008/06/03 13:27 (external edit)