Function Reference

Alphabetical List

---

Parallel Code Execution Constructs for automatically running code in parallel Distributed and Codistributed Arrays Data partitioned across multiple MATLAB sessions Jobs and Tasks Parallel computation through individual tasks Interlab Communication Within a Communicating Job Communications between labs during job execution Graphics Processing Unit Transferring data and running code on the GPU Utilities Utilities for using Parallel Computing Toolbox

Back to Top

Parallel Code Execution

Parallel Code on a MATLAB Pool Parallel computations on a pool of MATLAB sessions Profiles, Input, and Output Data access and setup control Interactive Functions Parallel code development and debugging

Back to Top of Section

Parallel Code on a MATLAB Pool

batch	Run MATLAB script or function on worker
Composite	Create Composite object
distributed	Create distributed array from data in client workspace
matlabpool	Open or close pool of MATLAB sessions for parallel computation
parfor	Execute code loop in parallel
spmd	Execute code in parallel on MATLAB pool

Back to Top of Section

Profiles, Input, and Output

diary	Display or save Command Window text of batch job
exist	Check whether Composite is defined on labs
load	Load workspace variables from batch job
parallel.clusterProfiles	Names of all available cluster profiles
parallel.defaultClusterProfile	Examine or set default cluster profile
parallel.exportProfile	Export one or more profiles to file
parallel.importProfile	Import cluster profiles from file
pctRunOnAll	Run command on client and all workers in matlabpool
subsasgn	Subscripted assignment for Composite
subsref	Subscripted reference for Composite

Back to Top of Section

Interactive Functions

mpiprofile	Profile parallel communication and execution times
pmode	Interactive Parallel Command Window

Back to Top

Distributed and Codistributed Arrays

Toolbox Functions Array creation and manipulation Overloaded MATLAB Functions Predefined arrays with overloaded functions

Back to Top of Section

Toolbox Functions

codistributed	Create codistributed array from replicated local data
codistributed.build	Create codistributed array from distributed data
codistributed.colon	Distributed colon operation
codistributor	Create codistributor object for codistributed arrays
codistributor1d	Create 1-D codistributor object for codistributed arrays
codistributor1d.defaultPartition	Default partition for codistributed array
codistributor2dbc	Create 2-D block-cyclic codistributor object for codistributed arrays
codistributor2dbc.defaultLabGrid	Default computational grid for 2-D block-cyclic distributed arrays
for	for-loop over distributed range
gather	Transfer distributed array data or GPUArray to local workspace
getCodistributor	Codistributor object for existing codistributed array
getLocalPart	Local portion of codistributed array
globalIndices	Global indices for local part of codistributed array
isaUnderlying	True if distributed array's underlying elements are of specified class
isComplete	True if codistributor object is complete
isreplicated	True for replicated array
redistribute	Redistribute codistributed array with another distribution scheme

Back to Top of Section

Overloaded MATLAB Functions

codistributed.cell	Create codistributed cell array
codistributed.eye	Create codistributed identity matrix
codistributed.false	Create codistributed false array
codistributed.Inf	Create codistributed array of Inf values
codistributed.NaN	Create codistributed array of Not-a-Number values
codistributed.ones	Create codistributed array of ones
codistributed.rand	Create codistributed array of uniformly distributed pseudo-random numbers
codistributed.randn	Create codistributed array of normally distributed random values
codistributed.spalloc	Allocate space for sparse codistributed matrix
codistributed.speye	Create codistributed sparse identity matrix
codistributed.sprand	Create codistributed sparse array of uniformly distributed pseudo-random values
codistributed.sprandn	Create codistributed sparse array of uniformly distributed pseudo-random values
codistributed.true	Create codistributed true array
codistributed.zeros	Create codistributed array of zeros
distributed.cell	Create distributed cell array
distributed.eye	Create distributed identity matrix
distributed.false	Create distributed false array
distributed.Inf	Create distributed array of Inf values
distributed.NaN	Create distributed array of Not-a-Number values
distributed.ones	Create distributed array of ones
distributed.rand	Create distributed array of uniformly distributed pseudo-random numbers
distributed.randn	Create distributed array of normally distributed random values
distributed.spalloc	Allocate space for sparse distributed matrix
distributed.speye	Create distributed sparse identity matrix
distributed.sprand	Create distributed sparse array of uniformly distributed pseudo-random values
distributed.sprandn	Create distributed sparse array of normally distributed pseudo-random values
distributed.true	Create distributed true array
distributed.zeros	Create distributed array of zeros
sparse	Create sparse distributed or codistributed matrix

Back to Top

Jobs and Tasks

Job Creation Job and task definition Job Management Job and task execution Task Execution Information Information on the processes evaluating a task Object Control Parallel Computing Toolbox objects

Back to Top of Section

Job Creation

createJob	Create distributed or independent job
createMatlabPoolJob	Create MATLAB pool job
createParallelJob	Create parallel job object
createTask	Create new task in job
dfeval	Evaluate function using cluster
dfevalasync	Evaluate function asynchronously using cluster
findResource	Find available parallel computing resources
jobStartup	File for user-defined options to run when job starts
mpiLibConf	Location of MPI implementation
mpiSettings	Configure options for MPI communication
pctconfig	Configure settings for Parallel Computing Toolbox client session
poolStartup	File for user-defined options to run on each worker when MATLAB pool starts
setupForParallelExecution	Set options for submitting parallel jobs to scheduler
taskFinish	User-defined options to run on worker when task finishes
taskStartup	User-defined options to run on worker when task starts

Back to Top of Section

Job Management

cancel	Cancel job or task
changePassword	Prompt user to change job manager password or MJS password
clearLocalPassword	Delete local store of user's job manager password
demote	Demote job in cluster queue
destroy	Remove job or task object from parent and memory
findJob	Find job objects stored in scheduler
findTask	Task objects belonging to job object
getAllOutputArguments	Output arguments from evaluation of all tasks in job object
getDebugLog	Read output messages from job run in CJS cluster
getJobSchedulerData	Get specific user data for job on generic scheduler
pause	Pause job manager queue
promote	Promote job in MJS cluster queue
resume	Resume processing queue in job manager
setJobClusterData	Set specific user data for job on generic cluster
setJobSchedulerData	Set specific user data for job on generic scheduler
submit	Queue job in scheduler
wait	Wait for job to change state or for GPU calculation to complete
waitForState	Wait for object to change state

Back to Top of Section

Task Execution Information

getAttachedFilesFolder	Folder into which AttachedFiles are written
getCurrentCluster	Cluster object that submitted current task
getCurrentJob	Job object whose task is currently being evaluated
getCurrentJobmanager	Job manager object that scheduled current task
getCurrentTask	Task object currently being evaluated in this worker session
getCurrentWorker	Worker object currently running this session
getFileDependencyDir	Directory where FileDependencies are written on worker machine

Back to Top of Section

Object Control

clear	Remove objects from MATLAB workspace
get	Object properties
inspect	Open Property Inspector
length	Length of object array
methods	List functions of object class
set	Configure or display object properties
size	Size of object array

Back to Top

Interlab Communication Within a Communicating Job

gcat	Global concatenation
gop	Global operation across all labs
gplus	Global addition
labBarrier	Block execution until all labs reach this call
labBroadcast	Send data to all labs or receive data sent to all labs
labindex	Index of this lab
labProbe	Test to see if messages are ready to be received from other lab
labReceive	Receive data from another lab
labSend	Send data to another lab
labSendReceive	Simultaneously send data to and receive data from another lab
numlabs	Total number of labs operating in parallel on current job
pload	Load file into parallel session
psave	Save data from parallel job session

Back to Top

Graphics Processing Unit

arrayfun	Apply function to each element of array on GPU
bsxfun	Binary singleton expansion function for GPUArray
existsOnGPU	Determine if GPUArray or CUDAKernel is available on GPU
feval	Evaluate kernel on GPU
gather	Transfer distributed array data or GPUArray to local workspace
gpuArray	Create array on GPU
gpuDevice	Query or select GPU device
gpuDeviceCount	Number of GPU devices present
parallel.gpu.CUDAKernel	Create GPU CUDA kernel object from PTX and CU code
reset	Reset GPU device and clear its memory
setConstantMemory	Set some constant memory on GPU
wait	Wait for job to change state or for GPU calculation to complete

Back to Top

Utilities

help	Help for toolbox functions in Command Window
pctRunDeployedCleanup	Clean up after deployed parallel applications

Objects — Alphabetical List

Functions — Alphabetical List

© 1984-2012- The MathWorks, Inc.    -   Site Help   -   Patents   -   Trademarks   -   Privacy Policy   -   Preventing Piracy   -   RSS