Introduction to NCI Gadi
Synopsis: This course is to give experienced supercomputer users a quickstart guide for using the National Computational Infrastructure’s (NCI) High Performance Computing (HPC) cluster ‘Gadi’.
This material will be presented live but will also be available as online material and as a recorded presentation.
Prerequisites
- Know how to connect to an ssh server from a terminal client
- Be competent using the Unix/Linux command line
- Understand general HPC concepts
- Know how to submit jobs to a PBS Pro (or similar) job scheduler
Follow-on courses: This course is the first of a number of SIH Gadi courses. We intend to provide courses for running specific optimised workflows on Gadi, eg Trinity transcriptome assembly, whole genome sequence analysis, and more. Watch this space!
Course survey!
Please fill out our course survey before you leave!
Help us help you!
Schedule
| 00:00 | 1. Introduction and agenda | SIH introduction and workshop agenda |
| 00:05 | 2. About NCI Gadi | Software and workflows suitable for Gadi |
| 00:14 | 3. Accounting | Querying disk and KSU available per project |
| 00:22 | 4. PBS jobs and Gadi-specific commands | PBS jobs on Gadi and compute node local disk |
| 00:42 | 5. Data transfer | Transfer to, from and within Gadi |
| 00:48 | 6. Software |
Global versus local software
Make your own local software compatible with module commands |
| 00:53 | 7. Optimisation | Selecting the appropriate resources for your job to maximise efficiency |
| 01:18 | 8. Example parallel job | Run multi-node parallel jobs on Gadi |
| 01:48 | Finish |
The actual schedule may vary slightly depending on the topics and exercises chosen by the instructor.