Parallel Computing Toolbox 5.2
Latest Features
Version 5.2
Released: 1 Sep 2011Version 5.2, part of Release 2011b, includes the following enhancements:
- GPU computing enhancements: random number generation on GPUs, more GPU-enabled MATLAB functions, and improved
arrayfunfunction for GPU computing applications in MATLAB - Increase in the number of local workers (from 8 to 12) supported by Parallel Computing Toolbox and by MATLAB Compiler generated executables from parallel applications
inv,sort,meshgrid,ndgrid,linspace, andlogspacefunctions available for distributed arrays- Job monitor tool on MATLAB desktop for tracking jobs launched via Parallel Computing Toolbox
- Point-and-click launch of MATLAB scripts for execution in batch mode from MATLAB desktop
See the Release Notes for details.
Update Now
- Download the latest version if your software maintenance service is current.
- Purchase or request a renewal quote if your maintenance service has lapsed.
- Try the latest version by downloading or requesting trial software.
Version 5.1
Released: 8 Apr 2011Version 5.1, part of Release 2011a, includes the following enhancements:
- GPU array indexing,
ndgrid,conv, andfilterfunctions, and enhancements toarrayfunfunction - Distributed arrays with
arrayfunandreshapefunctions and 2-D block-cyclic distribution in matrix multiplication and concatenation - Utilization of up to 8 workers on local computers for MATLAB Compiler generated applications and components
- Visual indicator on MATLAB desktop to indicate status of interactive MATLAB pool sessions
- Admin Center feature that remotely starts and stops system services associated with MathWorks job manager on all computers in a cluster
See the Release Notes for details.
Version 5.0
Released: 3 Sep 2010Version 5.0, part of Release 2010b, includes the following enhancements:
- GPU array interface that enables numerous MATLAB functions to perform single- and double-precision computations on NVIDIA CUDA-enabled GPUs (v1.3 compute capability and above)
- Ability to load and execute GPU kernels based on CUDA API directly from MATLAB
- Several linear algebra functions, such as
qrandchol, enhanced to operate on distributed arrays - MathWorks job manager enhanced, including improved secure access and the ability to run jobs as a specified user
- Generic scheduler interface enhanced, improving support for computing environments with nonshared file systems
See the Release Notes for details.
Version 4.3
Released: 5 Mar 2010Version 4.3, part of Release 2010a, includes the following enhancements:
- Ability to load and save distributed arrays held across multiple workers from and to a single MAT-file
- Support for single and double-complex precision, and 2-D block-cyclic distribution in several functions for distributed arrays
- New command-line functionality that allows remote startup of MATLAB Distributed Computing Server processes on cluster machines from the desktop
See the Release Notes for details.
Version 4.2
Released: 4 Sep 2009Version 4.2, part of Release 2009b, includes the following enhancements:
- New interface for accessing and manipulating distributed arrays on a pool of MATLAB workers
- Upgraded algorithms and enhanced visualization support for distributed arrays
- Enhanced MATLAB pool API, allowing dynamic adding or updating of the
FileDependenciesproperty while the MATLAB client is connected to a pool of workers - Support for job templates in Windows HPC Server 2008, enabling easier implementation and compliance with cluster-use policies for MATLAB jobs
See the Release Notes for details.