Pre-Cyberinfrastructure Day 2019 Workshops

Join us for the 2019 Pre-Cyberinfrastructure Day (CI) Workshops at MU’s Memorial Union Stotler Lounge on Monday, May 6, 2019.

Who should attend Pre-CI Day?

Pre-CI Day Workshops are for faculty, research scientists, graduate students, post-docs, and research-support staff who use (or are interested in using) advanced computing technologies.

Why attend?

Pre-CI Day Workshops offer hands-on learning about technologies, plus expert resources. Workshops range from beginning to advanced topics.

Registration

Registration is now open! Please register for the pre-conference workshops. While this event is free, registration is encouraged.

What to Bring

Be sure to bring your laptop and a charger for these hands-on workshops.  Laptops and chargers will not be provided.

Parking

For anyone traveling from outside Columbia MO, please contact wassmanr@missouri.edu for parking instructions.

Schedule

May 6 (Monday) Memorial Union’s Stotler Lounge

Time Topic
9:00 a.m. – 5:00 p.m. Utilizing Intel Capabilities for HPC

Presenter: Intel

Morning Topics:

  • Intel Architecture Overview
  • Intel Compilers
  • Intel Distribution for Python
  • Intel Performance Libraries

11:45-12:45: Lunch Break

Afternoon Topics:

  • Intel Advisor
  • Flow Graph Analyzer
  • Roof Line Analysis
  • Intel VTune
  • Intel Inspector
  • Intel MPI
  • Intel Trace Analyzer
  •  Q&A
9:00 a.m. – 5:00 p.m. CUDA in C/C++ for Deep Learning 

Presenter: Mike Gosnell, Nvidia University Ambassador 

This workshop teaches the fundamental tools and techniques for accelerating C/C++ applications to run on massively parallel GPUs with CUDA. Work your way through dozens of hands-on coding exercises using a live, GPU-enabled development environment in the cloud. Learn how to write code to be executed by a GPU accelerator, configure code parallelization using the CUDA thread hierarchy, manage and optimize memory migration between the CPU and GPU accelerator, leverage command line and visual profilers to guide your work, and utilize concurrent streams for instruction-level parallelism. Finish by implementing the workflow that you have learned on a new task — accelerating a fully functional, but CPU-only, particle simulator for observable massive performance gains. At the end of the workshop,
you will have access to additional resources to create new GPU-accelerated applications on your own.

Prerequisites: Basic C/C++ competency, including familiarity with variable types, loops, conditional statements, functions, and array manipulations. No previous knowledge of CUDA programming is assumed.

9:00 a.m. – 5:00 p.m. HPC Carpentry Workshop

Presenter: Kansas City Federal Reserve Bank

Morning topics:

  • Why use a Cluster – Higher Performance Computing (HPC)
  • Connecting to the Cluster
  • Scripts – turning a set of commands into a program
  • Working on a Cluster
  • Scheduling and Submitting Jobs
  • Using Cluster Resources Effectively

Noon to 1 pm: Lunch Break

Afternoon topics:

  • Navigating the System
  • File Management – Create, edit, move or delete files
  • Wildcards and Pipes
  • Variables and Loops –  turning a command set into a program
  • Accessing Software Packages
  • Transferring Files to and from the Cluster
  • Tips and Tricks – Such as monitoring jobs and how to get jobs easily scheduled