2014 TACC Summer Supercomputing Institute

Monday, June 16 - Friday, June 20, 2014

Audience

This week-long workshop is appropriate for all levels of researchers, faculty, staff, and graduate students, from new users of advanced computing technologies, to those who have research projects requiring powerful computing, visualization, storage, or software. We encourage participation from Minority Serving Institutions, Hispanic Serving Institutions, and Historically Black Colleges and Universities.

  • Researchers across disciplines: Mathematics, Engineering, Physics, Astronomy, Astrophysics, Cosmology, Geology & Geophysics, Computer Sciences, Biosciences, Nanosciences...
  • Graduate and undergraduate students
  • Current TACC & XSEDE users
  • Industrial affiliates

Location

ROC 1.900
Texas Advanced Computing Center
J.J. Pickle Research Campus
10100 Burnet Rd.
Austin, TX 78758

Overview

The Institute will provide researchers with an intensive introduction to using TACC's computing resources. Senior TACC staff will deliver presentations and lead interactive lab sessions focused on using TACC's advanced computing resources and technology.

  • Stampede: Dell PowerEdge C8220 Cluster with Intel Xeon Phi coprocessors
  • Lonestar: Dell Linux Cluster
  • Maverick: HP/NVIDIA Interactive Visualization and Data Analytics System
  • Ranch: Petascale archival facility

On January 7, 2013 TACC deployed a new compute cluster, Stampede. Funded by the National Science Foundation, this new cluster provides the community with access to 2 PFlops of Intel based microprocessor power and 8 PFlops of Intel MIC (Many Integrated Core) architecture technology. During the Institute students will receive a description of the system and TACC staff will present a session on how to use the new MIC architecture.

Lectures and Labs: Senior TACC staff will deliver presentations and lead interactive laboratory sessions:

  • Obtaining access to TACC resources and services
  • Reviewing the hardware and software available on TACC resources
  • Developing parallel programs with OpenMP and MPI
  • Using visualization and data analysis software and systems
  • GPGPU programming
  • Using the Intel Xeon Phi coprocessor (MIC)

Applications Seminars: Leading computational researchers will discuss their work, including examples of how they are utilizing TACC's resources.

Consulting: During the Institute, TACC staff will be available to assist participants in applying the techniques and technologies covered in the Institute to their own applications.

Applying to Attend the Institute

Applications to attend the Institute must be submitted by Friday, May 2, 2014. Applicants will receive notification of the status of their application by Friday, May 9, 2014.

Fee for Accepted Applicants

A fee of $150 is due by Friday, May 30, 2014, for accepted applicants. The fee covers lunch, snacks, and course materials.

Refund Policy

Accepted applicants needing to cancel must do so by Friday, June 6, 2014, in order to receive a refund.

Accommodations

Institute attendees traveling to the training are responsible for arranging and paying for their own travel, daily expenses, and hotel accommodations. Accepted applicants will receive a list of suggested hotels.

 

Slides

Day 1

Welcome & Introduction

TACC Hardware Overview

Linux HPC Cluster Programming Environments

Lab 1: Stampede, Maverick User Environment

Introduction to Parallel Computing

Serial and Vector Optimization

Lab2: Serial and Vector Optimization

Day 2

Parallel Applications with OpenMP

Parallel Applications with MPI

Lab 3: MPI - C, MPI - Fortran, OpenMP

Lab 4: Hands on Introduction to R

Introduction to Data Management

Day 3

Introduction to Scientific Visualization

Introduction to Information Visualization

Lab 5: Visualization using Paraview

Lab 6: Visualization using VisIt

Parallel Visualization

Lab 7: Remote and Collaboritive Visualization

Day 4

Parallel I/O Using MPI & HDF5

Parallel Optimizaton for HPC

Lab 8: Parallel Optimization

Data Analysis on TACC Resources

Introduction to Intel Xeon Phi Coprocessors Part I

Introduction to Intel Xeon Phi Coprocessors Part II

Lab 8: Hands on with Intel Xeon Phi

Day 5

Debugging Parallel Applications

Debug Lab

PerfExpert

Hands on OpenMP / MPI Hybrid Computing

Lab 9: Hands on OpenMP / MPI Hybrid Computing

An Overview of Profiling

For more information about the TACC Summer Supercomputing Institute, please contact John Lockman, TACC Training Coordinator: jlockman@tacc.utexas.edu, or 512-471-4097.