Difference between revisions of "HPC:Queues"

From HPC wiki
Line 8: Line 8:
 
=== PMACS HPC Cluster Queues ===  
 
=== PMACS HPC Cluster Queues ===  
  
Beginning '''June 1st, 2015''' , only the following queues will be available for use:
+
The following queues will be available for the users, and GPU queues access will be granted specifically.  Some of the queues are reserved, like 'admin'.
  
 
1. '''normal (default)''' : Intended for non-interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 1000. No Run Limit.<br>   
 
1. '''normal (default)''' : Intended for non-interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 1000. No Run Limit.<br>   
 
2. '''interactive''' : Intended for interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 10. Run Limit: 8hrs.<br>
 
2. '''interactive''' : Intended for interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 10. Run Limit: 8hrs.<br>
3. '''denovo''' : Intended for big-memory jobs. The default reservations are 1 CPU core and 24 GB of RAM. Per user job limit: 32. Run Limit: 5days.<br>
+
3. '''denovo''' : Intended for big-memory jobs. The default reservations are 88 CPU core and 500 GB of RAM. Per user job limit: 96. Run Limit: 3days.<br>
  
 
<pre>
 
<pre>
 
$ bqueues
 
$ bqueues
 
QUEUE_NAME      PRIO STATUS          MAX JL/U JL/P JL/H NJOBS  PEND  RUN  SUSP
 
QUEUE_NAME      PRIO STATUS          MAX JL/U JL/P JL/H NJOBS  PEND  RUN  SUSP
normal          30  Open:Active      - 1000    -    -    0    0    0    0
+
cpd-dev          34  Open:Active      -  105    -    -    1    0    1    0
interactive      30  Open:Active      -    -    -    -    1     0    1     0
+
normal          30  Open:Active      - 1000    -    - 3702  1776  1926    0
denovo           30  Open:Active      -  32   -    -    0    0    0    0
+
interactive      30  Open:Active      -  10    -    -    56    0    56    0
 +
admin            30  Open:Active      -  10    -    -    0    0    0    0
 +
globus          30  Open:Active      -  10    -    -    0    0    0    0
 +
denovo          30  Open:Active      -  96    -    -    88     0   88     0
 +
cpd              30  Open:Active      -  320    -    -    5     0     5     0
 +
rhel9            30  Open:Active      -   80   -    -     4    0    4    0
 +
gpu              30  Open:Active      -  64   -   -    0     0     0    0
 +
gpu1            30  Open:Active      -  48    -    -  188    40  148     0
 +
gpuphy           30  Open:Active      -  128    -    -  2048  1792  256    0
 +
gpu2            30  Open:Active      -  24    -    -    0    0    0    0
 +
test            30  Open:Active      -  50   -    -    0    0    0    0
 
</pre>
 
</pre>
  

Revision as of 18:38, 25 September 2025

Other Pages

PMACS HPC Cluster Queues

The following queues will be available for the users, and GPU queues access will be granted specifically. Some of the queues are reserved, like 'admin'.

1. normal (default) : Intended for non-interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 1000. No Run Limit.
2. interactive : Intended for interactive jobs, the default reservations are 1 vCPU core and 6GB of RAM. Per user job limit: 10. Run Limit: 8hrs.
3. denovo : Intended for big-memory jobs. The default reservations are 88 CPU core and 500 GB of RAM. Per user job limit: 96. Run Limit: 3days.

$ bqueues
QUEUE_NAME      PRIO STATUS          MAX JL/U JL/P JL/H NJOBS  PEND   RUN  SUSP
cpd-dev          34  Open:Active       -  105    -    -     1     0     1     0
normal           30  Open:Active       - 1000    -    -  3702  1776  1926     0
interactive      30  Open:Active       -   10    -    -    56     0    56     0
admin            30  Open:Active       -   10    -    -     0     0     0     0
globus           30  Open:Active       -   10    -    -     0     0     0     0
denovo           30  Open:Active       -   96    -    -    88     0    88     0
cpd              30  Open:Active       -  320    -    -     5     0     5     0
rhel9            30  Open:Active       -   80    -    -     4     0     4     0
gpu              30  Open:Active       -   64    -    -     0     0     0     0
gpu1             30  Open:Active       -   48    -    -   188    40   148     0
gpuphy           30  Open:Active       -  128    -    -  2048  1792   256     0
gpu2             30  Open:Active       -   24    -    -     0     0     0     0
test             30  Open:Active       -   50    -    -     0     0     0     0

Example usage of LSF commands

Documented here are a few of the commonly used LSF commands. They are NOT intended for copy-paste purposes, but are intended to provide some guidelines on how these can be used.

Batch (non-interactive) Job submission

To run a job in batch mode:

$ bsub <script_name> 

Example:

$ bsub sh sleep.sh
Job <9990021> is submitted to default queue <normal>.

To run a job in batch mode that needs more RAM:

$ bsub -M <memory_required> <script_name> 

Example (job that requires 30GB of RAM):

$ bsub -M 30720 sh sleep.sh
Job <9990022> is submitted to default queue <normal>.