Parallel Computing Tutorial, Part 7: spmd - Parallel Code Beyond parfor

From the series: Parallel and GPU Computing Tutorials

Harald Brunnhofer, MathWorks

Execute code simultaneously on workers, access data on worker workspaces, and exchange data between workers using Parallel Computing Toolbox™ and MATLAB Distributed Computing Server™.

Product Focus

  • Parallel Computing Toolbox

Series: Parallel and GPU Computing Tutorials

Parallel Computing Tutorial, Part 1: Product Landscape  3:42
Get an overview of parallel computing products used in this tutorial series.

Parallel Computing Tutorial, Part 2: Prerequisites and Setting Up  3:19
Review hardware and product requirements for running the parallel programs demonstrated in Parallel Computing Toolbox™ tutorials.

Parallel Computing Tutorial, Part 3: Quick Success with parfor  3:41
Review an introductory parfor example using Parallel Computing Toolbox™.

Parallel Computing Tutorial, Part 4: Deeper Insights into Using parfor  3:49
Convert for-loops to parfor-loops, and learn about factors governing the speedup of parfor-loops using Parallel Computing Toolbox™.

Parallel Computing Tutorial, Part 5: Batch Processing  6:38
Offload serial and parallel programs using batch command, and use the Job Monitor.

Parallel Computing Tutorial, Part 6: Scaling to Clusters and Cloud  6:21
Learn about considerations for using a cluster, creating cluster profiles, and running code on a cluster with MATLAB Distributed Computing Server™.

Parallel Computing Tutorial, Part 7: spmd - Parallel Code Beyond parfor  4:01
Execute code simultaneously on workers, access data on worker workspaces, and exchange data between workers using Parallel Computing Toolbox™ and MATLAB Distributed Computing Server™.

Parallel Computing Tutorial, Part 8: Distributed Arrays  5:43
Perform matrix math on very large matrices using distributed arrays in Parallel Computing Toolbox™.

Parallel Computing Tutorial, Part 9: GPU Computing with MATLAB  6:14
Learn about using GPU-enabled MATLAB functions, executing NVIDIA® CUDA™ code from MATLAB®, and performance considerations.