Partners High Performance Computing Cluster
Partners applications via Mac/PC
Clinical & research applications
DFCI bioinformatics computer
PHS Research Computing cluster
Bioinformatics news
Data Storage & Backup
Sharing files & collaboration
HIPAA, ePHI and research (internal)
RPDR
HPCGG
Biomedical Engineering Model Shop
Harvard's GForge Implementation
Institutional research distribution lists

 
pHPC account registration pHPC user guide pHPC services pHPC web protal



Working interactively with your job in cluster

Mostly PHS clusters take batch jobs and run them in the "background", users do not need to interact with the job during the execution. However, sometimes users do need to interact with the application, for example, the application need the input from command line or wait for mouse event in xwindows. In order to interact with your application, you need to submit jobs with flag "-I", and you need to separate the allocation from the interactive work. In the first step, submit your allocation request to LSF-HPC with an interactive shell as the job, in the second step, go to the allocated resource, do your real work.

The following example assumes you will request 1 computing core, and it will run an individual interactive job.

Step 1: 

[testy@n137 ~]$ bsub -I -n 1 -x -q interact /bin/bash
Job <668932> is submitted to default queue .
Waiting for dispatch ...
Starting on <25>
Attention:
1. The queue for interactive job is "interact" queue.

2. The bsub command requests 1 computing core and it runs /bin/bash on the allocated node for you. If resources are not immediately available, the terminal pauses at "Waiting for dispatch..." When "Starting on " displays, the resources are allocated and /bin/bash is running.

Step 2:
"Starting on < n25 > " shows allocation is on node n25, you can open another terminal and ssh to n25 to work on your jobs. (leave the first terminal open) 
[testy@n137 ~]$ ssh n25


Since node n25 have multiple computing cores, it can take more jobs from other users. You can specify "-x" flag to let the queue system allocate the entire computing node exclusively, use "bsub -I -n1 -x /bin/bash" when you submit the job to request resource.

Note: If you need to run xwindows, simply add "-X" flag (capital "X") for xwindows forwarding when you use ssh.


WARNING: DO NOT bypass queue system and ssh directly to computing node at anytime, your job will be terminated immediately and your account will be suspended, thanks!


Search: