Change Location × Victoria, BC

    Recent Locations

      WestGrid Research Computing Winter School @ University of Victoria in Victoria


      • WestGrid Research Computing Winter School @ University of Victoria Photo #1
      1 of 1
      December 10, 2019

      Tuesday   9:00 AM

      Engineering / Computer Science (ECS) Building , Room 660
      Victoria, British Columbia

      Performers:
      • No Performers Listed
      EVENT DETAILS
      WestGrid Research Computing Winter School @ University of Victoria

      Similar to WestGrid's successful Research Computing Summer Schools, this Winter School will provide hands-on introductions to essential tools and skills used in computational research.  Researchers in all disciplines are invited to participate. No prior experience is required for any of the topics being covered.  These courses are open to anyone -- researchers from other institutions and organizations outside of the University of Victoria are welcome to register. TENTATIVE PROGRAM: Introduction to bash command line (1/2 day, with Alex Razoumov) This session will be a hands-on introduction to the Linux command line and the interaction with a remote server. We will review basic Linux commands, file management (edit, copy, remove and remote-transfer files), directories and the file system, remote access, and bash scripting. Introduction to HPC: using clusters to speed up your research (1/2 day, with Alex Razoumov and Belaid Moa) An introduction to high-performance computing for those without any HPC experience. We will start with an overview of the HPC cluster hardware and a quick description of the resources available on Compute Canada's national systems (Cedar, Graham, Béluga, Niagara). We will then take a look at the basic tools and techniques for working on a cluster: software environment and modules, overview of installed programming languages and compilers, quick introduction to parallel programming frameworks, working with make and Makefiles, and installing new software locally. Next, we will discuss 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 Python and R on the clusters, as well as best practices in cluster workflows. Introduction to scientific visualization: plotting, interactive 3D, scripting, animation, large datasets (full day, with Alex Razoumov) We will start with a brief introduction to scientific visualization and will spend the first hour doing simple 1D/2D/3D plotting using *plot.ly* open-source Python library. The rest of the day we will study 3D scientific visualization with ParaView, an open source, multi-platform data analysis and visualization tool designed to run on a variety of hardware from an individual laptop to large supercomputers. With ParaView users can interactively visualize 2D and 3D data sets defined on structured, adaptive and unstructured meshes or particles, animate these datasets in time, and manipulate them with a variety of filters. ParaView supports both interactive (GUI) and scripted (including offscreen) visualization, and is an easy and fun tool to learn. Virtualize and Dockerize (full day, with Belaid Moa) The clouds are becoming an integral part of ARC initiatives. Therefore, Compute Canada has adopted Openstack clouds to accommodate the increasing demand for running ARC tools and technologies that the traditional high-performance clusters cannot handle. In this workshop, we will discuss when to seek cloud resources and how to access them. We will also introduce the Openstack middleware and show how researchers use it to embrace cloud as infrastructure. Attendees will be able to launch their virtual machines with different operating systems such as Centos, Ubuntu, and Windows. They will be able to create persistent storage and attach it to their instances. We will also introduce the Openstack CLI interface so that users can script their repetitive cloud tasks. While doing so, we will address the heavyweight aspects of virtualization and introduce attendees to the amazing world of containers, especially dockers. We will learn how to create dockers and launch them on cloud instances. We will also show how an application can be dockerized and pushed to a docker repository for reuse and deployment. We also show how complex applications can be decomposed into microservices, each handled by an independent and self-contained docker. Lastly, we will show how to run your dockerized application on our HPC clusters via Singularity. Foundations of parallel programming: introduction to the Chapel programming language (full day, with Alex Razoumov) This course is a general introduction to the main concepts of parallel programming, using Chapel programming language to illustrate the basic concepts and ideas. Chapel is a relatively new language for both shared- and distributed-memory programming, with easy-to-use, high-level abstractions for both task and data parallelism that make it ideal for learning parallel programming for a novice HPC user. Chapel is incredibly intuitive, striving to merge the ease-of-use of Python and the performance of traditional compiled languages such as C and Fortran. Parallel constructs that typically take tens of lines of MPI code can be expressed in only a few lines of Chapel code. Chapel is open source and can run on any Unix-like operating system, with hardware support from laptops to large HPC systems. DAILY SCHEDULE: Each day will follow the same schedule: Morning: 9:00 am - 12:00 pm (noon) Lunch Break: 12:00 - 1:00 pm(attendees responsible for their own lunch) Afternoon: 1:00 - 4:00 pm All sessions will be held at the University of Victoria, Engineering/Computer Science (ECS) Building, Room 660.  COST: For a single fee, you can attend as many sessions as you like. This fee covers the cost of light refreshments each day provided for coffee breaks in the morning and afternoon. If you have any dietary restrictions, please contact us. Registration Fees: Faculty, student or staff @ University of Victoria or another WestGrid Member Institution: $30 Faculty, student or staff @ non-Member academic institution: $40 Public / industry / non-Academic: $50   Note: If you require assistance with covering the cost of this event, or are unable to pay by credit card, please contact us and alternative arrangements can be made. QUESTIONS: Email WestGrid Training & Visualization Coordinator, Alex Razoumov: alex.razoumov@westgrid.ca.

      Categories: University & Alumni | Science | Technology

      Event details may change at any time, always check with the event organizer when planning to attend this event or purchase tickets.