This shows you the differences between two versions of the page.
Both sides previous revision Previous revision Next revision | Previous revision | ||
tutorial:torque [2016/10/07 09:22] sertalpbilal Moved everything |
tutorial:torque [2024/02/28 13:12] (current) mjm519 [Table] |
||
---|---|---|---|
Line 2: | Line 2: | ||
TORQUE provides control over batch jobs and distributed computing resources. It is an advanced open-source product based on the original PBS project and incorporates the best of both community and professional development. It incorporates significant advances in the areas of scalability, | TORQUE provides control over batch jobs and distributed computing resources. It is an advanced open-source product based on the original PBS project and incorporates the best of both community and professional development. It incorporates significant advances in the areas of scalability, | ||
+ | |||
+ | |||
+ | |||
+ | ===== Prerequisite ===== | ||
+ | In order to extract your output and error results in Torque, you need to have password-less connection between nodes. If you have not set it once, execute the following commands. These commands create a public and private key so that when a node want to transfer a file to your home folder, it does not require the password. | ||
+ | After connecting to polyps enter: | ||
+ | |||
+ | <code bash> | ||
+ | ssh-keygen -N "" | ||
+ | </ | ||
+ | |||
+ | Then just press ENTER for any question. After that type the following commands: | ||
+ | |||
+ | <code bash> | ||
+ | touch ~/ | ||
+ | chmod 600 ~/ | ||
+ | cat ~/ | ||
+ | </ | ||
+ | Now, you will get the error log and output log files for your jobs. | ||
+ | |||
+ | |||
+ | |||
===== Hardware ===== | ===== Hardware ===== | ||
Line 9: | Line 31: | ||
| polyp1--polyp15 | | polyp1--polyp15 | ||
| polyp30 | 24 Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz | 128 GB | 2x K80 (4GPUs) | | | polyp30 | 24 Intel(R) Xeon(R) CPU E5-2620 v3 @ 2.40GHz | 128 GB | 2x K80 (4GPUs) | | ||
+ | |||
+ | |||
+ | Configured Resources as provided in the Maui scheduler. This is pulled from Torque: | ||
+ | PROCS: 16 | ||
+ | MEM: 31G | ||
+ | SWAP: 63G | ||
===== Submitting Jobs ===== | ===== Submitting Jobs ===== | ||
Line 22: | Line 50: | ||
#PBS -o / | #PBS -o / | ||
#PBS -l nodes=1: | #PBS -l nodes=1: | ||
+ | #PBS -l pmem=2GB: | ||
#PBS -q batch | #PBS -q batch | ||
Line 39: | Line 68: | ||
</ | </ | ||
If you do not want to write the submission script you can do it just by calling | If you do not want to write the submission script you can do it just by calling | ||
- | < | + | < |
Now, we will run the code but we are setting the job parameters using '' | Now, we will run the code but we are setting the job parameters using '' | ||
- | ===== Important | + | ===== Options ===== |
+ | ^ Option | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
+ | | '' | ||
- | * '' | + | You can find detailed information |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | * '' | + | |
- | + | ||
- | See [[http:// | + | |
+ | <note tip>You need to use option '' | ||
===== Monitoring and Removing jobs ===== | ===== Monitoring and Removing jobs ===== | ||
To show the jobs use '' | To show the jobs use '' | ||
< | < | ||
- | To show jobs of some user use '' | + | To show jobs of some user use '' |
+ | <code shell> | ||
+ | qdel JOB_ID | ||
+ | </ | ||
+ | Moreover, you can use the following command: | ||
+ | < | ||
+ | < | ||
+ | < | ||
==== Queues ==== | ==== Queues ==== | ||
- | We have few queues '' | + | We have few queues '' |
< | < | ||
- | Queue | + | Queue |
- | ---------------- ------ -------- -------- ---- --- --- -- ----- | + | ---------------- |
- | medium | + | MOSEK |
- | short -- -- | + | AMPL |
- | long | + | long 30 1 |
- | batch | + | gpu 4 0 |
- | verylong | + | verylong |
- | ----- ----- | + | medium |
- | 0 0 | + | coraverylong |
+ | special | ||
+ | batch | ||
+ | short 0 | ||
+ | urgent | ||
+ | background | ||
+ | mediumlong | ||
</ | </ | ||
+ | |||
+ | If you want to use AMPL or MOSEK, you have to use queue: AMPL or MOSEK, because we have limited licenses for them. | ||
+ | |||
+ | |||
You can see limits using this command '' | You can see limits using this command '' | ||
- | ^ Queue ^ Wall Time ^ | + | ^ Queue |
- | | batch | 01: | + | | urgent |
- | | short | 02: | + | | batch | 01: |
- | | medium | 04: | + | | short |
- | | long | 72: | + | | medium |
- | | very long | + | | mediumlong |
+ | | long | 72: | ||
+ | | verylong | ||
+ | | special | ||
+ | | background | ||
+ | | gpu | ||
+ | | AMPL | | 200 | 8 | 6 | ||
+ | | MOSEK | ||
+ | |||
+ | |||
+ | |||
+ | Notes: | ||
+ | * Urgent queue has no limits and jobs have a higher priority over all other jobs in the queues. Please be respectful of others if using this queue to complete time sensitive or critical jobs. | ||
+ | * background queue has no limits and jobs have a lower priority over all other jobs in the queues. | ||
+ | ===== Examples ===== | ||
+ | |||
+ | ==== Submitting a Small or Large Memory Job ==== | ||
+ | |||
+ | You can use the option '' | ||
+ | |||
+ | <code bash limited.sh> | ||
+ | qsub -l pmem=4gb, | ||
+ | </ | ||
- | ==== Examples ==== | + | Sometimes your job needs more memory. You can choose a larger memory size with the same option: |
- | === Submitting Large Memory Job === | + | <code bash large.pbs> |
- | Sometimes your job needs more memory. This can be achieved | + | To see what resources have been assigned |
- | < | + | < |
+ | <code bash ulimit> | ||
+ | core file size (blocks, -c) 0 | ||
+ | data seg size | ||
+ | scheduling priority | ||
+ | file size | ||
+ | pending signals | ||
+ | max locked memory | ||
+ | max memory size | ||
+ | open files (-n) 65536 | ||
+ | pipe size (512 bytes, -p) 8 | ||
+ | POSIX message queues | ||
+ | real-time priority | ||
+ | stack size (kbytes, -s) unlimited | ||
+ | cpu time | ||
+ | max user processes | ||
+ | virtual memory | ||
+ | file locks (-x) unlimited</ | ||
- | === Running MATLAB | + | **[[https:// |
+ | ==== Running MATLAB | ||
You just have to create a submission job which looks like this | You just have to create a submission job which looks like this | ||
Line 109: | Line 196: | ||
#PBS -o / | #PBS -o / | ||
#PBS -l nodes=1: | #PBS -l nodes=1: | ||
+ | #PBS -l pmem=2GB: | ||
#PBS -q batch | #PBS -q batch | ||
Line 114: | Line 202: | ||
</ | </ | ||
- | === Interactive Jobs === | + | <note tip>Use **-singleCompThread** [[https:// |
+ | ==== Running Solvers ==== | ||
+ | |||
+ | In order to run solvers (such as Gurobi/ | ||
+ | |||
+ | < | ||
+ | |||
+ | This flag enables the solver to find necessary authentication information. | ||
+ | |||
+ | ==== Interactive Jobs ==== | ||
If you do not care where you run your job just use '' | If you do not care where you run your job just use '' | ||
Line 124: | Line 221: | ||
and you will be running interactive session on polyp15. | and you will be running interactive session on polyp15. | ||
- | === Using GPU's === | + | ==== Using GPU' |
Line 132: | Line 229: | ||
However, first you have to have a permission to use GPU (given by Prof. Takac) -- this is just formality to allow to certain users to use video driver on polyp30 | However, first you have to have a permission to use GPU (given by Prof. Takac) -- this is just formality to allow to certain users to use video driver on polyp30 | ||
- | === Running MPI and Parallel Jobs === | + | If you are using TensorFlow in Python, you can set the limit on amount of GPU memory using: |
+ | < | ||
+ | config_tf.gpu_options.per_process_gpu_memory_fraction = p</ | ||
+ | in which **//p//** is the percent of GPU memory (a number between zero and one). | ||
+ | |||
+ | ==== Running MPI and Parallel Jobs ==== | ||
<code bash mpi.pbs> | <code bash mpi.pbs> | ||
Line 154: | Line 256: | ||
</ | </ | ||
- | Allocating more than one CPU under PBS can be done in a number of ways, using the -l flag and the following resource descriptions: | + | Allocating more than one CPU under PBS can be done in a number of ways, using the '' |
* nodes - specifies the number of separate nodes that should be allocated | * nodes - specifies the number of separate nodes that should be allocated | ||
Line 160: | Line 262: | ||
* ppn - how many processes to allocate for each node | * ppn - how many processes to allocate for each node | ||
- | The allocation made by pbs will be reflected in the contents of the nodefile, which can be accessed via the $PBS_NODEFILE environment variable. | + | The allocation made by pbs will be reflected in the contents of the nodefile, which can be accessed via the '' |
The difference between ncpus and ppn is a bit subtle. ppn is used when you actually want to allocate multiple processes per node. ncpus is used to qualify the sort of nodes you want, and only secondarily to allocate multiple slots on a cpus. Some examples should help. | The difference between ncpus and ppn is a bit subtle. ppn is used when you actually want to allocate multiple processes per node. ncpus is used to qualify the sort of nodes you want, and only secondarily to allocate multiple slots on a cpus. Some examples should help. | ||
Line 195: | Line 297: | ||
</ | </ | ||
- | ==== ADVANCED | + | ===== Mass Operations ===== |
+ | |||
+ | ==== Submitting multiple jobs ==== | ||
+ | An easy way to submit multiple jobs via PBS is using a batch script. Suppose we would like to give all file names inside a folder with MPS extension into our solver. We can write a PBS Script such as | ||
+ | <code bash submit.pbs> | ||
+ | cd / | ||
+ | / | ||
+ | </ | ||
+ | and a BASH script: | ||
+ | <code bash bashloop.sh> | ||
+ | for f in dataset/ | ||
+ | do | ||
+ | qsub -q batch -v FILENAME=$f submit.pbs | ||
+ | done | ||
+ | </ | ||
+ | Here, option '' | ||
+ | |||
+ | After having these two files, simply calling | ||
+ | < | ||
+ | ./ | ||
+ | </ | ||
+ | will submit all jobs into Torque. | ||
+ | |||
+ | ==== Cancelling all jobs ==== | ||
+ | You can call | ||
+ | <code bash> | ||
+ | qselect -u < | ||
+ | </ | ||
+ | to cancel all of your running jobs. | ||
+ | |||
+ | <code bash> | ||
+ | qselect -u < | ||
+ | </ | ||
+ | will cancel all jobs (both running jobs and queue). | ||
+ | |||
+ | |||
+ | ===== Advanced ===== | ||
Line 220: | Line 358: | ||
+ | ==== Tensorflow with GPU ==== | ||
+ | To use tensorflow with a specific GPU, say GPU 1, you can simply set | ||
+ | <code bash> | ||
+ | export CUDA_VISIBLE_DEVICES=1 | ||
+ | </ | ||
+ | and then schedule your jobs with Torque to perform experiments on GPU 1. | ||
+ | |||
+ | |||
+ | ====== MOAB Scheduler ====== | ||
+ | PBS Torque is used to schedule and run jobs on our cluster. Two PBS processes are required to run jobs. On the PBS server, the pbs_server process runs to accept your job and add it to the queue. It will also dispatch the job to the nodes to run under the pbs_mom process. | ||
+ | |||
+ | |||
+ | ==== Useful MOAB Commands ==== | ||
+ | 1. [[https:// | ||
+ | |||
+ | 2. [[https:// | ||
+ | |||
+ | 3. [[https:// | ||
+ | |||
+ | ====Useful External Resources==== | ||
+ | [[https:// | ||
+ | [[https:// | ||
+ | [[http:// | ||
+ | [[https:// | ||