HPC/Applications/matlab/Parallel - Revision history

Stern at 22:40, February 18, 2014

2014-02-18T22:40:35Z

← Older revision		Revision as of 22:40, February 18, 2014
Line 28:		Line 28:
	*: Same, plus: Discontinued Support for <code>parallel.cluster.Mpiexec</code>		*: Same, plus: Discontinued Support for <code>parallel.cluster.Mpiexec</code>
	*: Automatic start of a parallel pool when executing code that uses parfor or spmd		*: Automatic start of a parallel pool when executing code that uses parfor or spmd

			=== Terminology ===
			; Batch Job:
			: This is a little relation to jobs executed on Carbon under PBS. MATLAB has its own, peculiar, [http://www.mathworks.com/help/releases/R2013a/distcomp/batch-processing.html Batch Processing] engine that allows the user to run code in the background of a MATLAB interactive session, while having full use of that session (command prompt etc.).
			; Cluster Profile:
			: A definition for a parallel environment, typically either for running on a local machine or on Carbon compute nodes.

	== Getting started ==		== Getting started ==
Line 33:		Line 39:
	* http://www.mathworks.com/help/releases/R2013a/distcomp/getting-started-with-parallel-computing-toolbox.html		* http://www.mathworks.com/help/releases/R2013a/distcomp/getting-started-with-parallel-computing-toolbox.html
	* http://www.mathworks.com/help/releases/R2013a/distcomp/introduction-to-parallel-solutions.html		* http://www.mathworks.com/help/releases/R2013a/distcomp/introduction-to-parallel-solutions.html
			* http://www.mathworks.com/help/releases/R2013a/distcomp/parallel-configurations-for-cluster-access.html

	== Caveats using MDCS ==		== Caveats using MDCS ==

Stern: /* Getting started */

2014-02-18T22:32:22Z

Getting started

← Older revision		Revision as of 22:32, February 18, 2014
Line 30:		Line 30:

	== Getting started ==		== Getting started ==
	~~{{Tag~~:~~todo}}~~		; TODO:
			* http://www.mathworks.com/help/releases/R2013a/distcomp/getting-started-with-parallel-computing-toolbox.html
			* http://www.mathworks.com/help/releases/R2013a/distcomp/introduction-to-parallel-solutions.html

	== Caveats using MDCS ==		== Caveats using MDCS ==

Stern at 22:29, February 18, 2014

2014-02-18T22:29:21Z

← Older revision		Revision as of 22:29, February 18, 2014
Line 1:		Line 1:
	== Introduction ==		== Introduction ==
	MathWorks offers two components to operate MATLAB in parallel.		MathWorks offers two components to operate MATLAB in parallel.
	# [http://www.mathworks.com/help/distcomp/ ~~The~~ Parallel Computing Toolbox] (PCT).		# The [http://www.mathworks.com/help/distcomp/ '''Parallel Computing Toolbox'''] (PCT).
	#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.		#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.
	# [http://www.mathworks.com/products/distriben/ ~~The~~ MATLAB Distributed Computing Server] (MDCS, formerly known as MDCE).		#* http://www.mathworks.com/help/distcomp/ (current release)
			#* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)
			# The [http://www.mathworks.com/products/distriben/ '''MATLAB Distributed Computing Server'''] (MDCS, formerly known as MDCE).
	#: Requires and extends the Toolbox for use across ''multiple machines''.		#: Requires and extends the Toolbox for use across ''multiple machines''.

Line 16:		Line 18:

	=== Note on MATLAB releases ===		=== Note on MATLAB releases ===
	* http://www.mathworks.com/help/distcomp/ (current release)
	* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)

	The MATLAB version that is installed on Carbon as of 2014-02 is R2013a, the last release supported under Enterprise Linux 5.x.		The MATLAB version that is installed on Carbon as of 2014-02 is R2013a, the last release supported under Enterprise Linux 5.x.
	Keep this in mind when navigating MathWorks documentation.		Keep this in mind when navigating MathWorks documentation.
Line 30:		Line 29:
	*: Automatic start of a parallel pool when executing code that uses parfor or spmd		*: Automatic start of a parallel pool when executing code that uses parfor or spmd

	== Caveats ==		== Getting started ==
			{{Tag:todo}}

			== Caveats using MDCS ==
			* At the close of an MDCS <code>matlabpool</code>, Carbon will send a job completion mail like this:
	An error has occurred processing your job, see below.		An error has occurred processing your job, see below.
	Post job file processing error; job 435379.sched1.carboncluster on host n189/7+n189/6+…+n187/0		Post job file processing error; job 435379.sched1.carboncluster on host n189/7+n189/6+…+n187/0
Line 39:		Line 42:
	*** end error output		*** end error output
	Output retained on that host in: /var/spool/torque/undelivered/''435379''.sched1.carboncluster.OU		Output retained on that host in: /var/spool/torque/undelivered/''435379''.sched1.carboncluster.OU
			: This mail can be ignored.

Stern at 22:23, February 18, 2014

2014-02-18T22:23:32Z

← Older revision		Revision as of 22:23, February 18, 2014
Line 1:		Line 1:
	== Introduction ==		== Introduction ==
	MathWorks offers two components to operate MATLAB in parallel.		MathWorks offers two components to operate MATLAB in parallel.
	# [http://www.mathworks.com/help/~~releases~~/~~product-description.html~~ The Parallel Computing Toolbox] (PCT).		# [http://www.mathworks.com/help/distcomp/ The Parallel Computing Toolbox] (PCT).
	#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.		#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.
	# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] (MDCS, formerly known as MDCE).		# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] (MDCS, formerly known as MDCE).
Line 12:		Line 12:
	* can run parallel applications interactively or in batch.		* can run parallel applications interactively or in batch.

	Adding MDCS:		Adding MDCS,
	* applications using PCT can run on a cluster (across multiple nodes) without changing the code.		* applications using PCT can run on a cluster (across multiple nodes) without changing the code.

	== Note on MATLAB releases ==		=== Note on MATLAB releases ===
	* http://www.mathworks.com/help/distcomp/ (current release)		* http://www.mathworks.com/help/distcomp/ (current release)
	* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)		* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)

Stern at 22:22, February 18, 2014

2014-02-18T22:22:27Z

← Older revision		Revision as of 22:22, February 18, 2014
Line 3:		Line 3:
	# [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Toolbox] (PCT).		# [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Toolbox] (PCT).
	#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.		#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.
	# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] – MDCS, formerly known as MDCE.		# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] (MDCS, formerly known as MDCE).
	#: Requires and extends the Toolbox for use across ''multiple machines''.		#: Requires and extends the Toolbox for use across ''multiple machines''.

Stern at 22:22, February 18, 2014

2014-02-18T22:22:04Z

← Older revision		Revision as of 22:22, February 18, 2014
Line 7:		Line 7:

	The Parallel Computing Toolbox:		The Parallel Computing Toolbox:
	* ~~Provides~~ high-level constructs (such as parallel for-loops, special array types, and parallelized numerical algorithms) to parallelize applications without CUDA or MPI programming.		* provides high-level constructs (such as parallel for-loops, special array types, and parallelized numerical algorithms) to parallelize applications without CUDA or MPI programming,
	* ~~Supports~~ CUDA-enabled NVIDIA GPUs (available on Carbon login nodes and gen3 nodes).		* supports CUDA-enabled NVIDIA GPUs (available on Carbon login nodes and gen3 nodes),
	* ~~Supports~~ up to twelve workers (computational engines) to execute applications locally on a multicore machine.		* supports up to twelve workers (computational engines) to execute applications locally on a multicore machine,
	* ~~Can~~ run parallel applications interactively or in batch.		* can run parallel applications interactively or in batch.

	Adding MDCS,		Adding MDCS:
	* ~~Applications~~ using PCT can run on a cluster (across multiple nodes) without changing the code.		* applications using PCT can run on a cluster (across multiple nodes) without changing the code.

	== ~~Documentation~~ ==		== Note on MATLAB releases ==
	* http://www.mathworks.com/help/distcomp/ (current release)		* http://www.mathworks.com/help/distcomp/ (current release)
	* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)		* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)

Stern at 22:20, February 18, 2014

2014-02-18T22:20:59Z

← Older revision		Revision as of 22:20, February 18, 2014
Line 19:		Line 19:
	* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)		* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)

	The ~~PCT~~ version installed on Carbon as of 2014-02 is R2013a, the last release supported under Enterprise Linux 5.x.		The MATLAB version that is installed on Carbon as of 2014-02 is R2013a, the last release supported under Enterprise Linux 5.x.
	Keep this in mind when navigating MathWorks documentation.		Keep this in mind when navigating MathWorks documentation.
	There have been some changes from R2013b onward, but R2013a code constructs will continue to run for now.		There have been some changes from R2013b onward, but R2013a code constructs will continue to run for now.

Stern at 22:19, February 18, 2014

2014-02-18T22:19:44Z

Stern: /* Introduction */

2014-02-18T22:04:24Z

Introduction

← Older revision		Revision as of 22:04, February 18, 2014
Line 8:		Line 8:
	The Parallel Computing Toolbox:		The Parallel Computing Toolbox:
	* Provides high-level constructs (such as parallel for-loops, special array types, and parallelized numerical algorithms) to parallelize applications without CUDA or MPI programming.		* Provides high-level constructs (such as parallel for-loops, special array types, and parallelized numerical algorithms) to parallelize applications without CUDA or MPI programming.
			* Supports CUDA-enabled NVIDIA GPUs (available on Carbon login nodes and gen3 nodes).
	* Supports up to twelve workers (computational engines) to execute applications locally on a multicore machine.		* Supports up to twelve workers (computational engines) to execute applications locally on a multicore machine.
	* Can run parallel applications interactively or in batch.		* Can run parallel applications interactively or in batch.

Stern: Created page with "== Introduction == MathWorks offers two components to operate MATLAB in parallel. # [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Too..."

2014-02-18T22:01:38Z

Created page with "== Introduction == MathWorks offers two components to operate MATLAB in parallel. # [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Too..."

New page

== Introduction ==
MathWorks offers two components to operate MATLAB in parallel.
# [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Toolbox] (PCT).
#: Provides the language features and capabilities for execution on a ''single machine.''
# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] – MDCS, formerly known as MDCE.
#: Requires and extends the Toolbox for use across ''multiple machines''.

The Parallel Computing Toolbox:
* Provides high-level constructs (such as parallel for-loops, special array types, and parallelized numerical algorithms) to parallelize applications without CUDA or MPI programming.
* Supports up to twelve workers (computational engines) to execute applications locally on a multicore machine.
* Can run parallel applications interactively or in batch.

Under MDCS,
* Applications using PCT can run on a cluster (across multiple nodes) without changing the code.

== Concepts ==
* http://www.mathworks.com/help/distcomp/
* or http://www.mathworks.com/help/releases/R2013a/distcomp/
* http://www.mathworks.com/help/distcomp/release-notes.html?searchHighlight=matlabpool
*: parpool: New command-line interface (replaces matlabpool)
* http://www.mathworks.com/help/mdce/release-notes.html?searchHighlight=matlabpool
*: Same, plus: Discontinued Support for parallel.cluster.Mpiexec
*: Automatic start of a parallel pool when executing code that uses parfor or spmd

== Caveats ==
An error has occurred processing your job, see below.
Post job file processing error; job 435379.sched1.carboncluster on host n189/7+n189/6+…+n187/0

Unable to copy file /var/spool/torque/spool/''435379''.sched1.carboncluster.OU to /home/stern/''somedir''/Job1/Job1.log
*** error from copy
/bin/cp: cannot create regular file `/home/stern/''somedir''/Job1/Job1.log': No such file or directory
*** end error output
Output retained on that host in: /var/spool/torque/undelivered/''435379''.sched1.carboncluster.OU

← Older revision		Revision as of 22:19, February 18, 2014
Line 2:		Line 2:
	MathWorks offers two components to operate MATLAB in parallel.		MathWorks offers two components to operate MATLAB in parallel.
	# [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Toolbox] (PCT).		# [http://www.mathworks.com/help/releases/product-description.html The Parallel Computing Toolbox] (PCT).
	#: Provides the language features and capabilities for execution on a ''single machine.''		#: Provides the language features and capabilities for execution on a ''single machine'', including GPU use.
	# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] – MDCS, formerly known as MDCE.		# [http://www.mathworks.com/products/distriben/ The MATLAB Distributed Computing Server] – MDCS, formerly known as MDCE.
	#: Requires and extends the Toolbox for use across ''multiple machines''.		#: Requires and extends the Toolbox for use across ''multiple machines''.
Line 12:		Line 12:
	* Can run parallel applications interactively or in batch.		* Can run parallel applications interactively or in batch.

	~~Under~~ MDCS,		Adding MDCS,
	* Applications using PCT can run on a cluster (across multiple nodes) without changing the code.		* Applications using PCT can run on a cluster (across multiple nodes) without changing the code.

	== ~~Concepts~~ ==		== Documentation ==
	* http://www.mathworks.com/help/distcomp/		* http://www.mathworks.com/help/distcomp/ (current release)
	* or http://www.mathworks.com/help/releases/R2013a/distcomp/		* http://www.mathworks.com/help/releases/R2013a/distcomp/ (R2013a)

			The PCT version installed on Carbon as of 2014-02 is R2013a, the last release supported under Enterprise Linux 5.x.
			Keep this in mind when navigating MathWorks documentation.
			There have been some changes from R2013b onward, but R2013a code constructs will continue to run for now.

			For the record, the changes are:
	* http://www.mathworks.com/help/distcomp/release-notes.html?searchHighlight=matlabpool		* http://www.mathworks.com/help/distcomp/release-notes.html?searchHighlight=matlabpool
	*: parpool: New command-line interface (replaces matlabpool)		*: parpool: New command-line interface (replaces matlabpool)
	* http://www.mathworks.com/help/mdce/release-notes.html?searchHighlight=matlabpool		* http://www.mathworks.com/help/mdce/release-notes.html?searchHighlight=matlabpool
	*: Same, plus: Discontinued Support for parallel.cluster.Mpiexec		*: Same, plus: Discontinued Support for <code>parallel.cluster.Mpiexec</code>
	*: Automatic start of a parallel pool when executing code that uses parfor or spmd		*: Automatic start of a parallel pool when executing code that uses parfor or spmd