Hoffman2:Introduction - Revision history

Hwang: /* External Links / Notes */

2021-02-16T18:15:48Z

External Links / Notes

← Older revision		Revision as of 18:15, 16 February 2021
Line 113:		Line 113:
	==External Links / Notes==		==External Links / Notes==
	*[http://www.hoffman2.idre.ucla.edu/ Hoffman2 Webpage]		*[http://www.hoffman2.idre.ucla.edu/ Hoffman2 Webpage]
	*[~~http~~://www.hoffman2.idre.ucla.edu/~~usage~~-~~status/~~ Hoffman2 Statistics]		*[https://www.hoffman2.idre.ucla.edu/About/Utilization-graph.html Hoffman2 Statistics]
	*[http://www.hoffman2.idre.ucla.edu/data-storage Hoffman2 Data Storage]		<!--*[http://www.hoffman2.idre.ucla.edu/data-storage Hoffman2 Data Storage]
	*[http://www.hoffman2.idre.ucla.edu/computing Hoffman2 Computing]		*[http://www.hoffman2.idre.ucla.edu/computing Hoffman2 Computing] -->
	*[[Hoffman2:Introduction-Historical_Notes \| Historical Notes]]		*[[Hoffman2:Introduction-Historical_Notes \| Historical Notes]]

Dmargolis at 21:28, 14 November 2017

2017-11-14T21:28:10Z

← Older revision		Revision as of 21:28, 14 November 2017
Line 81:		Line 81:
	====Temporary Storage====		====Temporary Storage====
	When running a computing job on Hoffman2, reading and writing a bunch of files in your home directory can be slow. So faster temporary storage is available to use for ongoing jobs. Read the official description [http://www.hoffman2.idre.ucla.edu/data-storage/ here].		When running a computing job on Hoffman2, reading and writing a bunch of files in your home directory can be slow. So faster temporary storage is available to use for ongoing jobs. Read the official description [http://www.hoffman2.idre.ucla.edu/data-storage/ here].
			<!--
	=====work=====		=====work=====
	:'''/work'''		:'''/work'''
Line 89:		Line 89:

	:If your job reads from or writes to a file repeatedly, you may save time by keeping that file in this temporary directory and then moving it to your home directory at job completion so it is not deleted.		:If your job reads from or writes to a file repeatedly, you may save time by keeping that file in this temporary directory and then moving it to your home directory at job completion so it is not deleted.
			-->
	=====scratch=====		=====scratch=====
	:'''/u/scratch/[u]/[username]'''		:'''/u/scratch/[u]/[username]'''
Line 96:		Line 96:
	====Cold Storage====		====Cold Storage====
	Cold storage is the idea of storing things away for a long time. Disk space on hoffman2 can be very expensive.		Cold storage is the idea of storing things away for a long time. Disk space on hoffman2 can be very expensive.
	IDRE has a cloud archival storage service [~~http~~://~~www.cass.~~idre.ucla.edu/ here]		IDRE has a cloud archival storage service [https://idre.ucla.edu/cass/ here]

	===Univa Grid Engine===		===Univa Grid Engine===

Dmargolis at 21:15, 14 November 2017

2017-11-14T21:15:47Z

← Older revision		Revision as of 21:15, 14 November 2017
Line 11:		Line 11:
	* Univa Grid Engine		* Univa Grid Engine

	[[File:~~Hoffman2-layout~~.~~png~~]]		[[File:CCN_Hoffman2_Workshop.jpg]]
	<br/>		<br/>
	''Image taken from a previous ATS "Using Hoffman2 Cluster" slide deck''		''Image taken from a previous ATS "Using Hoffman2 Cluster" slide deck''

Dmargolis at 21:09, 14 November 2017

2017-11-14T21:09:44Z

← Older revision		Revision as of 21:09, 14 November 2017
Line 2:		Line 2:

	==What is Hoffman2?==		==What is Hoffman2?==
	The Hoffman2 Cluster is a campus computing resource at UCLA and is named for Paul Hoffman (1947-2003). It is maintained by the [https://idre.ucla.edu/ IDRE] at UCLA and the main official webage is [http://www.hoffman2.idre.ucla.edu/ here]. With many high end processor, data storage, and backup technologies, it is a useful tool for executing research computations especially when working with large datasets. More than 1000 users are currently registered and the cluster sees tremendous usage. Click [[Hoffman2:Getting an Account\|here]] to find out how to join. In September 2014 alone, there were more than 5.5 million compute hours logged. See more usage statistics [~~http~~://www.hoffman2.idre.ucla.edu/~~usage-~~status/ here].		The Hoffman2 Cluster is a campus computing resource at UCLA and is named for Paul Hoffman (1947-2003). It is maintained by the [https://idre.ucla.edu/ IDRE] at UCLA and the main official webage is [http://www.hoffman2.idre.ucla.edu/ here]. With many high end processor, data storage, and backup technologies, it is a useful tool for executing research computations especially when working with large datasets. More than 1000 users are currently registered and the cluster sees tremendous usage. Click [[Hoffman2:Getting an Account\|here]] to find out how to join. In September 2014 alone, there were more than 5.5 million compute hours logged. See more usage statistics [https://www.hoffman2.idre.ucla.edu/status/ here].

	==Anatomy of the Computing Cluster==		==Anatomy of the Computing Cluster==
Line 9:		Line 9:
	* Computing Nodes		* Computing Nodes
	* Storage Space		* Storage Space
	* ~~Sun~~ Grid Engine ~~(a brain of sorts)~~		* Univa Grid Engine

	[[File:Hoffman2-layout.png]]		[[File:Hoffman2-layout.png]]
	<br/>		<br/>
	''Image taken from a previous ATS "Using Hoffman2 Cluster" slide deck ~~and modified for our point.~~''		''Image taken from a previous ATS "Using Hoffman2 Cluster" slide deck''


Line 99:		Line 99:

	===Univa Grid Engine===		===Univa Grid Engine===
	The ~~Sun~~ Grid Engine is the brains behind how jobs get executed on the cluster. When you request that a script be run on Hoffman2, the UGE looks at the resources you requested (how much memory, how many computing cores, how many computing hours, etc) and puts your job in a queue (a waiting line for those not familiar with British English) based on your requirements. Less demanding jobs generally get front loaded while more demanding ones must wait for adequate resources to free up. The UGE tries to schedule jobs on computing nodes in order to make the most efficient use of the resources available.		The Univa Grid Engine is the brains behind how jobs get executed on the cluster. When you request that a script be run on Hoffman2, the UGE looks at the resources you requested (how much memory, how many computing cores, how many computing hours, etc) and puts your job in a queue (a waiting line for those not familiar with British English) based on your requirements. Less demanding jobs generally get front loaded while more demanding ones must wait for adequate resources to free up. The UGE tries to schedule jobs on computing nodes in order to make the most efficient use of the resources available.

	====Queues====		====Queues====

Acho: /* External Links / Notes */

2016-04-18T21:36:10Z

External Links / Notes

← Older revision		Revision as of 21:36, 18 April 2016
Line 112:		Line 112:

	==External Links / Notes==		==External Links / Notes==
	*[http://~~hpc~~.ucla.edu/~~hoffman2/hoffman2.php~~ Hoffman2 Webpage]		*[http://www.hoffman2.idre.ucla.edu/ Hoffman2 Webpage]
	*[http://~~hpc~~.ucla.edu/~~hoffman2/h2stat~~/~~h2stat.php~~ Hoffman2 Statistics]		*[http://www.hoffman2.idre.ucla.edu/usage-status/ Hoffman2 Statistics]
	*[http://~~hpc~~.ucla.edu~~/hoffman2/data-storage~~/data-storage~~.php~~ Hoffman2 Data Storage]		*[http://www.hoffman2.idre.ucla.edu/data-storage Hoffman2 Data Storage]
	*[http://~~hpc~~.ucla.edu~~/hoffman2/computing~~/computing~~.php~~ Hoffman2 Computing]		*[http://www.hoffman2.idre.ucla.edu/computing Hoffman2 Computing]
	*[[Hoffman2:Introduction-Historical_Notes \| Historical Notes]]		*[[Hoffman2:Introduction-Historical_Notes \| Historical Notes]]

Acho: /* Queues */

2016-04-18T21:34:04Z

Queues

← Older revision		Revision as of 21:34, 18 April 2016
Line 107:		Line 107:
	; highp		; highp
	: For jobs requesting at most 14 days of computing time. These are required to run on nodes owned by your group.		: For jobs requesting at most 14 days of computing time. These are required to run on nodes owned by your group.
	And there are others. Read about them [http://~~hpc~~.ucla.edu~~/hoffman2~~/computing/~~computing.php~~ here].		And there are others. Read about them [http://www.hoffman2.idre.ucla.edu/computing/ here].

	Find out how to submit computing jobs to the [[Hoffman2:Submitting Jobs\| Hoffman2 Cluster.]]		Find out how to submit computing jobs to the [[Hoffman2:Submitting Jobs\| Hoffman2 Cluster.]]

Acho: /* Queues */

2016-04-18T21:33:23Z

Queues

← Older revision		Revision as of 21:33, 18 April 2016
Line 103:		Line 103:
	====Queues====		====Queues====
	There is more than one queue on Hoffman2. Each is for a slightly different purpose:		There is more than one queue on Hoffman2. Each is for a slightly different purpose:
	~~; express~~
	~~: For jobs requesting at most 2 hours of computing time.~~
	; interactive		; interactive
	: For jobs requesting at most 24 hours of computing time and requiring the ability for users to interact with the program running.		: For jobs requesting at most 24 hours of computing time and requiring the ability for users to interact with the program running.
Line 112:		Line 110:

	Find out how to submit computing jobs to the [[Hoffman2:Submitting Jobs\| Hoffman2 Cluster.]]		Find out how to submit computing jobs to the [[Hoffman2:Submitting Jobs\| Hoffman2 Cluster.]]


	==External Links / Notes==		==External Links / Notes==

Acho: /* Temporary Storage */

2016-04-18T21:33:04Z

Temporary Storage

← Older revision		Revision as of 21:33, 18 April 2016
Line 80:		Line 80:

	====Temporary Storage====		====Temporary Storage====
	When running a computing job on Hoffman2, reading and writing a bunch of files in your home directory can be slow. So faster temporary storage is available to use for ongoing jobs. Read the official description [http://~~hpc~~.ucla.edu~~/hoffman2~~/data-storage/~~data-storage.php~~ here].		When running a computing job on Hoffman2, reading and writing a bunch of files in your home directory can be slow. So faster temporary storage is available to use for ongoing jobs. Read the official description [http://www.hoffman2.idre.ucla.edu/data-storage/ here].

	=====work=====		=====work=====
Line 93:		Line 93:
	:'''/u/scratch/[u]/[username]'''		:'''/u/scratch/[u]/[username]'''
	:Where ''[username]'' is replaced with your Hoffman2 username and ''[u]'' is replaced with the first letter of your username. Data here is accessible on all login and computing nodes. You can use up to 2TB of space here, but '''Any files older than 7 days may be automatically deleted by the system'''. Use the [[Hoffman2:UNIX Tutorial#Environment Variables\|UNIX environment variable]] <code>$SCRATCH</code> to reliably access your personal scratch directory.		:Where ''[username]'' is replaced with your Hoffman2 username and ''[u]'' is replaced with the first letter of your username. Data here is accessible on all login and computing nodes. You can use up to 2TB of space here, but '''Any files older than 7 days may be automatically deleted by the system'''. Use the [[Hoffman2:UNIX Tutorial#Environment Variables\|UNIX environment variable]] <code>$SCRATCH</code> to reliably access your personal scratch directory.


	====Cold Storage====		====Cold Storage====

Acho: /* Storage Space */

2016-04-18T21:26:22Z

Storage Space

← Older revision		Revision as of 21:26, 18 April 2016
Line 39:		Line 39:

	===Storage Space===		===Storage Space===
	For official and up-to-date information about storage space, [http://~~hpc~~.ucla.edu~~/hoffman2~~/data-storage/~~data-storage.php~~ click here]. If you want a quick overview, see below.		For official and up-to-date information about storage space, [http://www.hoffman2.idre.ucla.edu/data-storage/ click here]. If you want a quick overview, see below.

	====Long Term Storage====		====Long Term Storage====
	IDRE maintains high end storage systems (BlueArc and Panasas) for Hoffman2 disk space. There have built in redundancies and are fault tolerant. Redundant Backups are also available.		IDRE maintains high end storage systems (BlueArc and Panasas) for Hoffman2 disk space. There have built in redundancies and are fault tolerant. Redundant Backups are also available.

	If all of that sounded Greek to you, the important thing to understand is that there is a lot of disk space on Hoffman2 and IDRE takes great pains to make sure your data is safe. If you are paranoid, there is alternative [ backups ].		If all of that sounded Greek to you, the important thing to understand is that there is a lot of disk space on Hoffman2 and IDRE takes great pains to make sure your data is safe. If you are paranoid, there is alternative backups.

	=====Home Directories=====		=====Home Directories=====
Line 98:		Line 98:
	Cold storage is the idea of storing things away for a long time. Disk space on hoffman2 can be very expensive.		Cold storage is the idea of storing things away for a long time. Disk space on hoffman2 can be very expensive.
	IDRE has a cloud archival storage service [http://www.cass.idre.ucla.edu/ here]		IDRE has a cloud archival storage service [http://www.cass.idre.ucla.edu/ here]


	===Univa Grid Engine===		===Univa Grid Engine===

Acho: /* Computing Nodes */

2016-04-18T21:16:52Z

Computing Nodes

← Older revision		Revision as of 21:16, 18 April 2016
Line 27:		Line 27:
	As of April 2014, Hoffman2 is made up of more than 12,000 processors across three data centers and this number continues to grow as the cluster is expanded. The individual cores of the processors are where your programs gets executed when you submit a job to the cluster. There are ways to request different amount of resources, such as how much RAM or CPU cores your program/job needs.		As of April 2014, Hoffman2 is made up of more than 12,000 processors across three data centers and this number continues to grow as the cluster is expanded. The individual cores of the processors are where your programs gets executed when you submit a job to the cluster. There are ways to request different amount of resources, such as how much RAM or CPU cores your program/job needs.

	There is also a GPU cluster that has more than 300 nodes, but access to this must be requested separately from a normal Hoffman2 account. For more information, go [http://~~hpc~~.ucla.edu~~/hoffman2~~/computing/gpuq~~.php~~ here]		There is also a GPU cluster that has more than 300 nodes, but access to this must be requested separately from a normal Hoffman2 account. For more information, go [http://www.hoffman2.idre.ucla.edu/computing/gpuq/ here]

	The reason the number of computing cores continues to grow is because more resource groups (like individual research labs) join Hoffman2 and buy nodes to be integrated into the cluster. Nodes contributed by a resource group are guaranteed to that resource group and can be used to run longer jobs ([http://~~hpc~~.ucla.edu~~/hoffman2~~/computing/policies~~.php#highp~~ up to 14 days]). As of June 2013, the Cohen and Bookheimer groups on Hoffman2 have 96 cores:		The reason the number of computing cores continues to grow is because more resource groups (like individual research labs) join Hoffman2 and buy nodes to be integrated into the cluster. Nodes contributed by a resource group are guaranteed to that resource group and can be used to run longer jobs ([http://www.hoffman2.idre.ucla.edu/computing/policies/ up to 14 days]). As of June 2013, the Cohen and Bookheimer groups on Hoffman2 have 96 cores:
	* 6 nodes (installed pre 2010) each with		* 6 nodes (installed pre 2010) each with
	** 8 cores		** 8 cores