Workshops

Leading up to the conference, HPCS is hosting two days of HPC-related tutorials and workshops. These hands-on sessions are intended to be interactive, with practical exercises and activities scheduled within the workshop, so all participants are encouraged to bring their own laptops. If you have any questions related to the workshops, please contact us.

SUNDAY, APRIL 29

High-Performance Storage Systems
Full-Day Workshop
8:30 am – 4:30 pm
(Room 1520 – Barrick Gold)

This workshop will focus on technical knowledge transfer and user experiences in the area of high-performance storage systems, such as the Lustre file system, and storage area networks. No academic papers will be part of the session. Instead, the workshop will be a combination of technical training, user’s group-like sharing of experiences, and panel discussions.

Sage Advice for Scientific Computing
Half-Day Workshop
8:30 am – 12:30 pm
(Room 1600 – Canfor Policy)

There are many tools available for scientific computing. Sage provides a central wrapper giving you access to a large number of these from a unified interface. This tutorial will start with an overview of Sage and the tools available. Participants will then be taken through a number of scientific problems from the initial description of a problem, through analysis, numerical simulation/solution, to the final results analysis. The tutorial will look at problems from several different areas of research.

Introduction to PETSc
Half-Day Workshop
1:00 – 5:00 pm
(Room 1600 – Canfor Policy)

This workshop will provide an overview of the PETSc library and some basic general usage, scalability, and numerical analysis gotchas. Practical examples of the package’s use within research will be shared, as well as other examples from applications involved in problem-solving software environments, optimization, and simulations.

MONDAY, APRIL 30

Introduction to Visualization Using Paraview
Half-Day Workshop
8:00 am – 12:30 pm
(Room 1600 – Canfor Policy)

ParaView is an open source, multi-platform data analysis and visualization application. It has a client-server architecture to facilitate remote visualization of datasets, and generates level of detail (LOD) models to maintain interactive framerates for large datasets. ParaView users can quickly build visualizations to analyze their data using qualitative and quantitative techniques. The data exploration can be done interactively in 3D or programmatically using ParaView’s batch processing capabilities using Python. This workshop will explore basics of scientific visualization with 1-, 2- and 3-dimensional data sets through hands-on exercises.

Practical Introduction to Parallel I/O
Half-Day Workshop
8:00 am – 12:30 pm
(Room 1520 – Barrick Gold)

Large-scale computation can be much easier than reading or recording the results of the computation. This half-day workshop will discuss, with hands-on activities, the basics of three technologies for truly Parallel I/O, including MPI-IO, HDF5, and NetCDF4.

Introduction to GPU Computing Using CUDA
Half-Day Workshop
1:00 – 5:00 pm
(Room 1600 – Canfor Policy)

CUDA is a parallel computing architecture that enables dramatic increases in computing performance for graphics, 3D content, video and other processing-intensive applications by harnessing the power of GPUs. This workshop will provide an introduction to CUDA, with opportunities for hands-on demonstrations and activities. Participants are encouraged to be familiar with C/C++ development.

Gaussian-09 Best Practices
Half-Day Workshop
1:00 – 5:00 pm
(Room 1520 – Barrick Gold)

Gaussian 09, a product of Gaussian Inc., is the industry-standard code for performing of electronic structure modelling for many areas of Chemistry, Biology and Material Sciences. It is a comprehensive program package that incorporates a rich variety of Quantum Chemical theories, which often have diverse requirements for computational resources. This workshop will provide an overview of various issues of the Gaussian 09 code on Compute Canada high-performance computing systems, such as the availability of the code on selected machines, as well as licensing policies and procedures of obtaining access to the code. This workshop will also be used to discuss best practices of using Gaussian for frequently used types of computations: molecular Density Functional Theory, correlated wavefunction methods such as Coupled Clusters and Gx-theories, and periodic computations.