ARC 2024 Summer School Program: 8-11 July, 2024

Introduction to High-Performance Computing and Data Storage a

In the age of big data and complex computational challenges, the ability to efficiently process and store vast amounts of data is crucial across many disciplines. This workshop is designed for UBC researchers who are interested in using high-performance computing (HPC) and large-scale data storage solutions to facilitate their research projects. Participants will gain hands-on experience and deepen their understanding of the latest technologies and methodologies in HPC and data storage.

Course Objectives:  

• Understand the Fundamentals: Learn the basic principles of high-performance computing and how it differs from traditional computing. Understand the architecture and functioning of object data storage systems
• Explore HPC Systems: Dive into the components of HPC systems, including CPUs, GPUs, and networking considerations, and how they are integrated to perform complex computations efficiently
• Data Management Strategies: Gain insights into effective data storage, management, and retrieval techniques suitable for handling large datasets
• Hands-On Simulations: Engage in practical sessions where you will use HPC clusters to run simulations, process data, and optimize computational tasks

Target Audience:  UBC researchers involved in data-intensive projects requiring robust computing and storage solutions, but have little experience in high-performance computing. 

Prerequisite:  Basic knowledge of computer systems and data processing is recommended.

Instructors:  Jerry Li, Michael Nesrallah, Roman Baranowski

SWNG 221

Risky Research - Securing Your Secrets and Yourself b

Have you ever been stumped by questions on applications regarding your research’s risk for ethics, data access and funding and wondered how you can address these?  Join us for this three-part workshop and feel confident in responding to these questions.

• Part 1:  Researchers are frequently asked to classify their data in order to understand their obligations for its protection.  However, many find this a difficult task.  Part 1 is an interactive session where participants classify their data based on its riskiness, identify risk in their own research setting and determine controls to decrease the risk of malicious actors stealing valuable research resources and data
• Part 2:  Everyone has secrets.  In information security, these are your passwords, keys and certificates that provide access to valuable research infrastructure and data.  Part 2 is also an interactive session where participants learn how to securely manage their secrets to prevent unauthorized access to research systems and data
• Part 3:  In part 3, the workshop will round up other controls and strategies to improve cybersecurity hygiene both in the research setting and beyond

Course Objectives:  

• Identify risk in one’s research project or research environment
• Classify research data based on risk
• Securely manage secrets (passwords, certificates and keys)
• Understand the importance of cybersecurity in data management planning
• Adopt at least 1 cybersecurity hygiene strategy.

Target Audience:  UBC Researchers and students involved in research and research support

Please note:  This session runs from 1:00 - 4:00 PM.

Prerequisite:  None

Instructors:  Jessica Galo, Luc Letarte, Scott Baker

SWNG 221

Introduction to REDCap Flex c

REDCap Flex is a powerful digital data collection platform that can not only facilitate the collection and organization of your research data, but can help manage teams’ access and permission, automate the emailing of survey invitations, allow participants to “e-consent”, create reports to monitor data as it’s being collected, allow survey respondents to switch between languages, and much more. In this two-hour workshop we’ll cover the basics of how to set up a REDCap project in the first-half before moving into more advanced topics in the second.

Course Objectives:

• Understand how to get started with REDCap Flex and how to setup a new project
• Know the difference between a basic form and a survey, between a longitudinal and non-longitudinal project, and between a public survey link and a custom link
• Know the most important User Rights and how they ensure data privacy and security
• Be able to pipe fields and use calculations 

Target Audience:  UBC researchers of all levels who collect behavioural and/or medical data. 

Please note:  This session runs from 1:00 - 4:00 PM.

Prerequisite:  None

Instructor:  Gabriel Smith

SWNG 222

Building Containers to Simplify Software Stacks on HPC and Cloud – Sessions I & II d

Session I

Join us for the first part of an all-day workshop on creating software stacks in a containerized environment. For the morning we will be exploring the basics of containers and will utilize Spack to help build a simple software stack within a container. We will also look at what is needed to write a definition file to create a new container. 

Course Objective:

• Understand the basics of containers and how to run software with them
• Learn how to construct your own containers with Spack and from scratch
• Be able to utilize containers on HPC systems for research workflows

Target Audience:
UBC researchers and graduate students who are new to high-performance computing but looking to acquire practical skills in working with containers for software environments. Participants are expected to have basic knowledge of Linux command-line operations. No prior experience with containers or HPC systems is necessary. 

Prerequisite for Session I:  None

Session II

Join us for part two of an all-day workshop on creating software stacks in a containerized environment. In the afternoon session we will begin to cover more advanced topics. This session will look at three common advanced cases for HPC: Virtual Environment, GPU software, and multi-node workflows using MPI.

Course Objectives:  

• Be fluent in constructing containers to replace virtual environment
• Understand how to set up containers with GPU access
• Experience with multi-node jobs using MPI and containers

Target Audience:
UBC researchers and graduate students who are new to high-performance computing but looking to acquire practical skills in working with containers for software environments. 

Prerequisite for Session II:  Basic knowledge of Linux command-line operations and prior experience with containers or Building Containers to Simplify Software Stacks on HPC and Cloud Session I

Instructors:  Jacob Boschee and Ken Bigelow

SWNG 221

Introduction to the Commercial Cloud e

Through the new UBC Hybrid Cloud Service, researchers can now access the commercial cloud without needing to set up their own accounts or billing. Platforms supported include Amazon Web Services (AWS) and Microsoft Azure, with support for Google Cloud Platform (GCP) coming soon. If you are interested in learning how this service can support your use of the public cloud (e.g., cost management, data security) in your research and how you can get started, this is the session for you.

Course Objectives:

• Learn how to access the public cloud via the UBC Hybrid Cloud Service
• Understand the benefits of using UBC Hybrid Cloud Service
• Estimate commercial cloud costs with the commercial cloud providers’ calculators
• Learn about some common research use cases

Target Audience:  UBC researchers interested in using the commercial cloud to advance their research. 

Prerequisite: None

Instructors: Sanja LeBlanc, Brent Dunington, Michael Tang

SWNG 222

Running Jobs on HPC – Sessions I & II f

Session I

Join us for an engaging two-hour workshop designed to introduce the fundamentals of using SLURM, the leading job scheduler, for managing batch jobs in a high-performance computing (HPC) environment. This workshop is ideal for participants who have little to no prior experience with HPC or SLURM but wish to take advantage of the system and the computing resources for their research projects.

Course Objectives:

• Understand the architecture and principles of HPC and SLURM
• Learn to write and manage basic batch job scripts using SLURM commands
• Gain practical skills in job submission, monitoring, and management
• Troubleshoot common issues in batch job processing

Target Audience:  UBC researchers and graduate students new to high-performance computing but looking to acquire practical skills in managing computational tasks with the system. 

Prerequisite for Session I:  Basic knowldge of Linux command-line operations; no experience with SLURM or HPC required.

Session II

Designed for a diverse group of professionals, researchers, and students, this workshop offers an invaluable opportunity to learn how to utilize HPC resources more effectively. Participants will gain hands-on experience with array jobs, explore the depths of MPI for intricate parallel processing tasks, and unlock the powerful capabilities of GPUs for computationally demanding applications. Whether you're a scientist running complex simulations, a data analyst dealing with big data, or an IT professional managing HPC infrastructures, this workshop will equip you with the practical skills and knowledge to elevate your computational projects. Engage with expert instructors, collaborate with peers, and step into the world of high-performance parallel computing to transform the way you process and analyze data. Join us to harness the power of HPC and advance your computational endeavors!

Course Objectives:

• To provide participants with a comprehensive understanding and hands-on experience in executing and managing array jobs, developing MPI applications for parallel processing, and leveraging GPUs for high-performance computing tasks

Target Audience:  Researchers and Academics interested in conducting complex computational research that requires HPC resources.

Prerequisite for Session II: Basic understanding of programming (e.g., Python), Linux command line usage, and general concepts in Computer Science.

Instructors: Jerry Li, Jacob Boschee, Roman Baranowski

SWNG 221

Introduction to Digital Research Alliance of Canada g

The Digital Research Alliance of Canada (“the Alliance”) is a non-profit funded by the Government of Canada that serves Canadian researchers, with the objective of advancing Canada’s position as a leader in the knowledge economy on the international stage.

In the context of Advanced Research Computing (ARC), the Alliance provides high performance compute (HPC), storage, and cloud computing services free of charge to UBC researchers. 

In this session, we will explore the various HPC and cloud computing resources available via the Alliance, how to gain access to and use these resources for your research projects, and how to apply for the annual Resource Allocation Competition (RAC). 

Prerequisite: None

Instructor: Venkat Mahadevan

SWNG 222

Introduction to Cloud and Virtual Machines h

What is Cloud? What are Virtual Machines? Join us in the morning for an introduction of the cloud and the Digital Research Alliance of Canada community cloud. In this session, you will learn about the resources available to Canadian researchers and how the cloud can help advance your research. In the afternoon, participants will have the opportunity to have hands-on experience launching virtual machines and other resources in the community cloud.

Course Objectives:

• Understand what cloud is and the resources available
• Aware of the resources available in the Digital Research Alliance of Canada community cloud and how to get started
• Gain hands-on experience launching and accessing virtual machines hosted in the community cloud

Target Audience: This workshop is ideal for participants who have little to no prior experience with the cloud and virtual machines.

Prerequisite:  None

Instructor:  Michael Tang

SWNG 221

Parallel Computing with MATLAB Parts I & II i

Part 1: 
Introduce parallel and distributed computing in MATLAB with a focus on speeding up application codes and offloading compute. By working through common scenarios and workflows using hands-on demos, you will gain a detailed understanding of the parallel constructs in MATLAB, their capabilities, and some of the common hurdles that you'll encounter when using them. 

Course Objectives:

• Multithreading vs multiprocessing
• When to use parfor vs parfeval constructs
• Creating data queues for data transfer
• Leveraging NVIDIA GPUs
• Parallelizing Simulink models
• Working with large data

Part II: 
During this hands-on workshop, we will show users how to best submit MATLAB jobs to an HPC cluster.

Course Objectives:

• How to configure MATLAB to submit remote jobs to the HPC cluster
• The job submission workflow
• Ways to tune job submissions to the HPC cluster
• How to optimize job submissions
• Troubleshooting job submission techniques
• Best practices for rehosting code onto the HPC cluster

Prerequisite:  None

Instructor:  Raymond Norris

SWNG 222