Torque - Revision history

Mbaney at 13:15, 12 June 2024

2024-06-12T13:15:04Z

← Older revision		Revision as of 13:15, 12 June 2024
Line 1:		Line 1:
	'''Refer to the [https://wiki.umiacs.umd.edu/~~cbcb-private~~/index.php/~~Slurm private Slurm wiki] or the [https://wiki.umiacs.umd.edu/umiacs/index.php~~/~~SLURM~~ UMIACS SLURM wiki] instead of this page for current information.'''		'''Refer to the [https://wiki.umiacs.umd.edu/umiacs/index.php/Nexus/CBCB UMIACS SLURM wiki] instead of this page for current information.'''

Mbaney at 20:35, 16 November 2018

2018-11-16T20:35:34Z

← Older revision		Revision as of 20:35, 16 November 2018
Line 1:		Line 1:
	'''Refer to the private Slurm wiki or the UMIACS SLURM wiki instead of this page for current information.'''		'''Refer to the [https://wiki.umiacs.umd.edu/cbcb-private/index.php/Slurm private Slurm wiki] or the [https://wiki.umiacs.umd.edu/umiacs/index.php/SLURM UMIACS SLURM wiki] instead of this page for current information.'''

Adric: Abolish page

2018-11-16T19:31:16Z

Abolish page

Show changes

Adric: Add notice

2018-11-15T16:18:35Z

Add notice

← Older revision		Revision as of 16:18, 15 November 2018
Line 1:		Line 1:
			'''Note: refer to the Slurm wiki instead of this page for more current information.'''

	==Getting Started==		==Getting Started==

Adric: Replace references to old submit nodes to the new ones.

2018-11-15T16:11:16Z

Replace references to old submit nodes to the new ones.

← Older revision		Revision as of 16:11, 15 November 2018
Line 3:		Line 3:
	Torque is a resource manager that interacts with another program called Maui which provides the scheduling for the cluster. To get started you will need to ensure that your [https://wiki.umiacs.umd.edu/umiacs/index.php/SSH#SSH_Keys_.28and_Passwordless_SSH.29 SSH keys] are setup for password-less SSH. In our Torque environments this is critical to delivering the error and output of your jobs back to where the job was submitted from. Please note that [https://wiki.umiacs.umd.edu/umiacs/index.php/Fairshare Fair Share] is enabled in this setup, with a historical scope of 12 hours.		Torque is a resource manager that interacts with another program called Maui which provides the scheduling for the cluster. To get started you will need to ensure that your [https://wiki.umiacs.umd.edu/umiacs/index.php/SSH#SSH_Keys_.28and_Passwordless_SSH.29 SSH keys] are setup for password-less SSH. In our Torque environments this is critical to delivering the error and output of your jobs back to where the job was submitted from. Please note that [https://wiki.umiacs.umd.edu/umiacs/index.php/Fairshare Fair Share] is enabled in this setup, with a historical scope of 12 hours.

	The hosts that you can submit from are any CBCB Workstation and the <tt>~~ibissub00~~.umiacs.umd.edu</tt> and <tt>~~ibissub01~~.umiacs.umd.edu</tt> nodes.		The hosts that you can submit from are any CBCB Workstation and the <tt>cbcbsub00.umiacs.umd.edu</tt> and <tt>cbcbsub01.umiacs.umd.edu</tt> nodes.

	Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.		Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.

Mbaney at 17:36, 6 June 2017

2017-06-06T17:36:57Z

← Older revision		Revision as of 17:36, 6 June 2017
Line 3:		Line 3:
	Torque is a resource manager that interacts with another program called Maui which provides the scheduling for the cluster. To get started you will need to ensure that your [https://wiki.umiacs.umd.edu/umiacs/index.php/SSH#SSH_Keys_.28and_Passwordless_SSH.29 SSH keys] are setup for password-less SSH. In our Torque environments this is critical to delivering the error and output of your jobs back to where the job was submitted from. Please note that [https://wiki.umiacs.umd.edu/umiacs/index.php/Fairshare Fair Share] is enabled in this setup, with a historical scope of 12 hours.		Torque is a resource manager that interacts with another program called Maui which provides the scheduling for the cluster. To get started you will need to ensure that your [https://wiki.umiacs.umd.edu/umiacs/index.php/SSH#SSH_Keys_.28and_Passwordless_SSH.29 SSH keys] are setup for password-less SSH. In our Torque environments this is critical to delivering the error and output of your jobs back to where the job was submitted from. Please note that [https://wiki.umiacs.umd.edu/umiacs/index.php/Fairshare Fair Share] is enabled in this setup, with a historical scope of 12 hours.

	The hosts that you can submit from are any CBCB Workstation and the <tt>ibissub00.umiacs.umd.edu</tt>, <tt>ibissub01~~.umiacs.umd.edu</tt>, and <tt>heronsub00~~.umiacs.umd.edu</tt> nodes.		The hosts that you can submit from are any CBCB Workstation and the <tt>ibissub00.umiacs.umd.edu</tt> and <tt>ibissub01.umiacs.umd.edu</tt> nodes.

	Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.		Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.
Line 19:		Line 19:
	** The xlarge queue is restricted to members of the group cbcbtorque. If you need to run large jobs please send mail to staff@umiacs.umd.edu		** The xlarge queue is restricted to members of the group cbcbtorque. If you need to run large jobs please send mail to staff@umiacs.umd.edu

	~~=== Red Hat 5 Queues ===~~

	~~These queues submit jobs to the RHEL 5 nodes.~~

	* old_workstation - default memory is 4GB (max 47GB), default walltime is 8 hours (max one week), allows up to 4 jobs per user concurrently
	* old_throughput - no interactive jobs - default memory is 4GB (max 36GB), default walltime is 4 hours (max 18 hours), allows up to 125 jobs per user concurrently
	* old_large - no interactive jobs -default memory is 32GB (max 120GB), default walltime is 24 hours (max 11 days), allows up to 3 jobs per user concurrently
	<nowiki>*</nowiki>You can list nodes with a specific property by running "pbsnodes :property" where property is the specific property you want to see		<nowiki>*</nowiki>You can list nodes with a specific property by running "pbsnodes :property" where property is the specific property you want to see

Pknut777 at 20:02, 25 June 2015

2015-06-25T20:02:24Z

← Older revision		Revision as of 20:02, 25 June 2015
Line 7:		Line 7:
	Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.		Now that you have that setup here are the queues that are available to users. Use <code>qstat -Q -f</code> to see the resource limits for each queue.

	=== Red Hat 5 Queues ===		=== Red Hat 7 Queues ===

	* default- default memory is 3GB (max 4GB), default walltime is 1 hour (max 1 hour), allows up to 16 jobs per user concurrently		* default- default memory is 3GB (max 4GB), default walltime is 1 hour (max 1 hour), allows up to 16 jobs per user concurrently

Tgray26: /* Red Hat 7 Queues */

2015-06-25T19:56:08Z

Red Hat 7 Queues

← Older revision		Revision as of 19:56, 25 June 2015
Line 19:		Line 19:
	** The xlarge queue is restricted to members of the group cbcbtorque. If you need to run large jobs please send mail to staff@umiacs.umd.edu		** The xlarge queue is restricted to members of the group cbcbtorque. If you need to run large jobs please send mail to staff@umiacs.umd.edu

	=== Red Hat 7 Queues ===		=== Red Hat 5 Queues ===

	These queues submit jobs to the RHEL 7 heron* nodes.		These queues submit jobs to the RHEL 5 nodes.

	* ~~new_workstation~~ - default memory is 4GB (max 47GB), default walltime is 8 hours (max one week), allows up to 4 jobs per user concurrently		* old_workstation - default memory is 4GB (max 47GB), default walltime is 8 hours (max one week), allows up to 4 jobs per user concurrently
	* ~~new_throughput~~ - no interactive jobs - default memory is 4GB (max 36GB), default walltime is 4 hours (max 18 hours), allows up to 125 jobs per user concurrently		* old_throughput - no interactive jobs - default memory is 4GB (max 36GB), default walltime is 4 hours (max 18 hours), allows up to 125 jobs per user concurrently
	* ~~new_hthroughput - no interactive jobs - default memory is 4gb (max 8GB), default walltime is 3 hours (max 6 hours), allows up to 300 jobs per user concurrently~~		* old_large - no interactive jobs -default memory is 32GB (max 120GB), default walltime is 24 hours (max 11 days), allows up to 3 jobs per user concurrently
	* new_large - no interactive jobs -default memory is 32GB (max 120GB), default walltime is 24 hours (max 11 days), allows up to 3 jobs per user concurrently
	<nowiki>*</nowiki>You can list nodes with a specific property by running "pbsnodes :property" where property is the specific property you want to see		<nowiki>*</nowiki>You can list nodes with a specific property by running "pbsnodes :property" where property is the specific property you want to see

Lmendelo: /* Getting Started */

2015-03-03T18:07:30Z

Getting Started

← Older revision		Revision as of 18:07, 3 March 2015
Line 158:		Line 158:
	gpus = 0		gpus = 0
	</pre></blockquote>		</pre></blockquote>

			==Using CBCB Modules with Torque==

			To use [[CBCB Software Modules]] with Torque, you will need to add these lines to your <code>~/.bashrc</code>:

			. /usr/share/Modules/init/bash
			. /etc/profile.d/ummodules.sh

	===Host Monitoring===		==Host Monitoring==
	http://ganglia.umiacs.umd.edu/ganglia/?c=cbcb_compute&m=load_one&r=hour&s=by%20name&hc=4&mc=2		http://ganglia.umiacs.umd.edu/ganglia/?c=cbcb_compute&m=load_one&r=hour&s=by%20name&hc=4&mc=2

Lmendelo: /* qsub */

2015-03-03T18:03:53Z

qsub

← Older revision		Revision as of 18:03, 3 March 2015
Line 72:		Line 72:
	Interactive jobs allow you to schedule interactive shell access on Torque-scheduled compute nodes. You can get an interactive session with the -I (upper case i) option,		Interactive jobs allow you to schedule interactive shell access on Torque-scheduled compute nodes. You can get an interactive session with the -I (upper case i) option,

	qsub -I		qsub -I

	Please note that only the "workstation", "shell", and "default" queues allow interactive jobs. If you require larger resource allocations than the queue defaults, the -l (lower case L) flag still applies.		Please note that only the "workstation", "shell", and "default" queues allow interactive jobs. If you require larger resource allocations than the queue defaults, the -l (lower case L) flag still applies.

			====Array Jobs====

			Array jobs let you submit the same script multiple times, each with a different setting for the environment variable <code>PBS_ARRAYID</code>:

			qsub -q throughput -t 0-999 my_script.sh

			Torque will run 1000 instances of <code>my_script.sh</code> with the environment variable <code>PBS_ARRAYID</code> set to the range of values specified by the <code>-t</code> argument. In this case <code>my_script.sh</code> will be executed once with <code>PBS_ARRAYID=0</code>, again with <code>PBS_ARRAYID=1</code>, etc.

			You can also specify comma separate values for <code>PBS_ARRAYID</code>:

			qsub -q throughput -t 0,3,9 my_script.sh

	===qstat===		===qstat===