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		 Previous revision
habrok:job_management:partitions [2023/03/22 12:22] – [Partitions] fokkehabrok:job_management:partitions [2023/07/11 07:03] (current) – ↷ Links adapted because of a move operation admin
Line 25: Line 25:
 |gpu               |GPU nodes                                           | 3 days |See [[habrok:advanced_job_management:running_jobs_on_gpus|this page]] for more information| |gpu               |GPU nodes                                           | 3 days |See [[habrok:advanced_job_management:running_jobs_on_gpus|this page]] for more information|
 |himem             |Big memory nodes with 4 TB of memory and 80 cores   |10 days |                                  |                                | |himem             |Big memory nodes with 4 TB of memory and 80 cores   |10 days |                                  |                                |
-|gelifes           |Nodes purchased by the GELIFES institute            |10 days |See [[.partitions:gelifes|this page]] for more information|+|gelifes           |Nodes purchased by the GELIFES institute            |10 days |See [[.partition_details:gelifes|this page]] for more information|
  
  
Line 43: Line 43:
 We can also help you in optimizing your code so that it can run faster. Please [[habrok:introduction:support|contact]] us if you need help. We can also help you in optimizing your code so that it can run faster. Please [[habrok:introduction:support|contact]] us if you need help.
  
-In order to alleviate the scheduling problems we limit long running jobs to only part of the cluster. For example for the standard nodes maximum half of the machines may be used for jobs taking more than 3 days. The precise settings depend on the partition, and may be changed to improve job scheduling.+In order to alleviate the scheduling problems we limit long running jobs to only part of the cluster. For example for the standard nodes maximum 80% of the machines may be used for jobs taking more than 3 days. The precise settings depend on the partition, and may be changed to improve job scheduling.
  
 ===== Limits on number of jobs ===== ===== Limits on number of jobs =====
  
-To prevent the scheduling system from being flooded with jobs, there are some limits in place that define how many jobs you are allowed to have in each partition at any time. For each partition, we distinguish between short, medium, and long jobs. The following table shows the limits on the number of jobs that you can submit to each partition and for each type of job: +To prevent the scheduling system from being flooded with jobs, there are some limits in place that define how many jobs you are allowed to have in the cluster at any time. 
- +
-**WiP** +
- +
-^Partition^^ regular ^^^ himem ^^^ gpu ^^short^vulture^ +
-|Job’s time limit      |≤ 1 day|1-3 days|> 3 days|≤ 1 day|1-3 days|> 3 days|≤ 1 hour|1-24 hours|> 24 hours|≤ 30 minutes|≤ 12 hours| +
-|Allowed number of jobs|2500   |1500    |500     |300    |200     |100     |1600      |800        |400       |1000        |500       |+
  
 Note that these limits only count the current number of (waiting and running) jobs. So if you reach the limit, you can submit new jobs after some other ones have finished.\\ Note that these limits only count the current number of (waiting and running) jobs. So if you reach the limit, you can submit new jobs after some other ones have finished.\\