This lesson introduces using the High Performance Computing with a Raspberry Pi cluster.
Prerequisites
Some basic experience of the Unix command line is expected. The material covered in the Software Carpentry Introduction to Unix Shell (http://swcarpentry.github.io/shell-novice/) is more than enough to cover this.
| Setup | Download files required for the lesson | |
| 00:00 | 1. Introduction to High Performance Computing | What does a High Performance Computing system consist of? |
| 00:10 | 2. Logging in to the cluster | How can do I login to the cluster? |
| 00:45 | 3. Filesystems and Storage |
Where can I store my data?
What is the difference between scratch and home filestore? |
| 01:30 | 4. Running Jobs with Slurm | How do I run a job with the Slurm scheduler? |
| 02:25 | 5. Profiling Single Core Performance | How do I find the portion of a code snipped that consumes the longest time? |
| 03:25 | 6. Parallel Estimation of Pi |
What are data parallel algorithms?
How can I estimate the yield of parallelization without writing one line of code? How do I use multiple cores on a computer in my program? |
| 04:00 | 7. Distributing computations among computers with MPI |
What issued the message passing interface (MPI)?
How do I exploit parallelism using the message passing interface (MPI)? |
| 04:55 | 8. HPC Best Practice | What are the best practices for using an HPC system? |
| 04:55 | Finish |
The actual schedule may vary slightly depending on the topics and exercises chosen by the instructor.