Introduction to HPC: using clusters to speed up your research

Description:

We will start with an overview of the hardware of common HPC clusters, and quick description of the resources available on National systems (Cedar and Graham), and locally at UofS (local cluster Plato, DataStore and Globus). We will then continue learning the basic tools and techniques to work on a cluster: software environment and modules, overview of installed programming languages and compilers, working with Make files and installing new software locally. Finally, we will take a look at Slurm job scheduler: why use it, fairshare and priority, submitting serial jobs and job arrays, submitting OpenMP / MPI / hybrid / GPU jobs, working inside interactive jobs, and tracking your job's memory usage. We will also take a quick look at working with common packages such as R, Python and Matlab on the clusters, as well as best practices in cluster workflows.

 

The workshop will be given by Dr. Alex Razoumov, Westgrid Training and Visualization Coordinator, Compute Canada; Sergiy Stepanenko, Advanced Computing Coordinator, University of Saskatchewan, and Dr. Juan Zuniga, University of Saskatchewan.

 

Pre-requisites/requirements:

In this workshop we assume all attendees have a working knowledge of the Linux command line (this material will be covered in day 1 of our workshop series, Moving your research from Windows-based desktops to Linux and command line in remote servers). All attendees will need to bring their laptops with wireless access and with a remote SSH client installed (on Windows we recommend the free edition of https://mobaxterm.mobatek.net/download.html; on Mac and Linux no need to install anything).

Dates/schedule:

Wednesday January 16th, 2019. From 9:00am to 12:00 pm and from 1:00 pm to 4:00 pm.

This workshop is part of a 4-workshop-series that will provide hands-on introductions to essential tools and skills used in computational research. The other workshops in the series include:

  • Moving your research from Windows-based desktops to Linux and command line in remote servers,
  • Scientific Visualization,
  • Foundations of Parallel Programming and the Chapel programming language.

You can attend single sessions or the entire series. There is no cost to attend but you must register online in advance.

Objectives:

See course description for a list of topics covered

Prerequisites:

  • See course description

Other Recommendations:

There are no additional recommendations with this course

Duration:

6 hour Course

Registration Fees:

U of S Students: $0.00
U of S Faculty: $0.00
U of S Staff: $0.00
Off Campus: $0.00

Required Software:

There is no required software for this course.

Resources:

There are no additional resources for this course.

NOTE: Classes may have limited registration. If an offering is designated as 'Full', please email Training Services in order that we may accommodate your training needs. Due to last minute withdrawls, this class is subject to cancellation on short notice. A customized training solution may be offered in its place.

Registration Information
You must be logged in to see registration information.