DokuWiki

User Tools

Site Tools

Trace:
cluster:189

Differences

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

Link to this comparison view

Both sides previous revision Previous revision
Next revision Both sides next revision
cluster:189 [2020/01/11 17:47]
hmeij07 		cluster:189 [2020/01/11 17:53]
hmeij07 [Funding Policy]
Line 20: Line 20:
 Several months later a pattern emerged.  The Provost would annually contribute $25K if the HPC user base raised $15K annually.  That would amount to $160K in 4 years enough for a hardware refresh or new hardware acquisition.  Finances also contributed $10K for maintenance such as failed disks, network switches, etc, but these funds do not "roll over". Use it or loose it. All funds start July 1st. Several months later a pattern emerged.  The Provost would annually contribute $25K if the HPC user base raised $15K annually.  That would amount to $160K in 4 years enough for a hardware refresh or new hardware acquisition.  Finances also contributed $10K for maintenance such as failed disks, network switches, etc, but these funds do not "roll over". Use it or loose it. All funds start July 1st.
  
 +In order for the HPC user base to raise $15K annually, CPU and GPU hourly usage was deployed. A dictionary is maintained listing PIs and their members (students majors, lab students, grads, phd candidates, collaborators, etc).  Each PI then quarterly contributes to the user fund based on  a scheme yieldingq $15K annually.
  
 +Here is 2019's queue usage [[cluster:188|2019 Queue Usage]] and 2019 contribution scheme.
  
-funding policy +Contribution Scheme for 01 July 2019 onwards\\ 
-priority access policy+Hours (K) - Rate ($/CPU Hour)\\ 
 +  * 0-5 = Free 
 +  * >5-25 = 0.03 
 +  * >25-125 = 0.006 
 +  * >125-625 = 0.0012 
 +  * >625-3125 = 0.00024 
 +  * >3125 = 0.000048 
 +A cpu usage of 3,125,000 hours/year would cost $ 2,400.00 \\
  
 +
 +priority access policy
 user base stats, annual meeting, spring reading week user base stats, annual meeting, spring reading week
 2019 queue usage stats link 2019 queue usage stats link
Line 30: Line 41:
 hpcc stats cpu cores, gpus, mem, hdd (rough) link to guide hpcc stats cpu cores, gpus, mem, hdd (rough) link to guide
 latest deployment: nvidia gpu cloud on premise (docker containers) link latest deployment: nvidia gpu cloud on premise (docker containers) link
- 
-Add funding model scheme 2016 2019 
-3x gpu vs cpu.  
 Script preempts nodes every 2 hours.  Script preempts nodes every 2 hours. 
  
cluster/189.txt · Last modified: 2024/02/12 16:47 by hmeij07

Page Tools

Except where otherwise noted, content on this wiki is licensed under the following license: CC Attribution-Share Alike 4.0 International
CC Attribution-Share Alike 4.0 International Donate Powered by PHP Valid HTML5 Valid CSS Driven by DokuWiki