There are two platforms on which to run the PCT: locally on a PC or remotely on a cluster. Which is more appropriate for you depends on accessibility and job requirements.
- Run locally on a multicore desktop or laptop.
If you have access to a multicore PC or laptop, this is suitable for code development or for applications that can take advantage of up to 12 cores (or 8 cores with R2011a or older versions).
- If you don’t have MATLAB installed on your machine, please check if you qualify to install it on your local machine.
- The PCT works on single and multicore computers. However, speed up depends on the number of physical cores.
- A worker is MATLAB’s term for a processor.
- If you are using MATLAB licensed to BU, your computer needs to either be physically on campus or, if accessing from off-campus (e.g., home), log in via VPN. Running MATLAB will trigger an automatic process to check out the appropriate MATLAB and PCT licenses. There is no action required by you.
- Run remotely on RCS’s Shared Computing Cluster (SCC).
This is the only way to run PCT jobs with more than 12 processors. With the appropriate batch submission procedure, you may also use this server to run jobs on a single node as if it is on a desktop or laptop. MATLAB calls this “local” and we will adopt this terminology throughout this tutorial. The SCC’s nodes have varying number of cores: 8, 12, or 16. However, “local” limits the core usage to the smaller of the assigned node’s maximum core count and 12. For example, if the runtime assigned node has 8 physical cores, you can only use up to 8 cores for parallel processing. On the other hand, if your job runs on a node with 16 cores, you can only use 12 cores as that is the maximum limit imposed by the MATLAB PCT license.
- To access the SCC requires RCS userid. If you don’t already have one, please check for eligibility.
- Before running MATLAB parallel tasks on SCC’s remote nodes for the first time, you must set up a MATLAB configuration file.
- A MATLAB PCT user may use up to 64 workers in a single job or in the aggregate from multiple concurrent jobs.
- The SCC login node has 4 processors. You can develop codes and run light, short-duration, parallel jobs on it.
Caveat: timing collection using tic/toc or other timing devices may be unreliable on the login node because its four processors are shared among all logged in users.