Skip to main content

HPE Cray EX Programming and Optimization with PBS Pro

H8PG3S

Table of Contents

Table of Contents

    Course ID

    H8PG3S

    Duration

    4 days

    Format

    ILT/VILT

    Overview

    In this course, you learn about the HPE Cray Supercomputing programming environment and how it is used on HPE Cray EX series systems. The programming environment consists of compilers, libraries, debuggers, and analysis tools that enable developers to efficiently use massively parallel supercomputers at scale. The HPE Cray EX supercomputer is the latest high-performance computing solution from HPE and is the platform for the world's first Exascale class system.

    Course ID

    H8PG3S

    Duration

    4 days

    Format

    ILT/VILT

    Audience

    This course is ideal for end users of HPE Cray EX series systems with the HPE Cray Supercomputing programming environment using PBS Pro for a workload manager. This course is also appropriate for HPE and customer support engineers who support end users of HPE Cray EX series systems with the HPE Cray Supercomputing programming environment.

    Prerequisites

    Before attending this course, you should have:

    Objectives

    After completing this course, you should be able to:

    • Provide an architectural overview of the HPE Cray EX supercomputer including the HPE Slingshot network
    • Preform basic Lustre configuration to optimize file I/O in their applications
    • List and describe the various components of the HPE Cray Supercomputing programming environment
    • Use the PBS Pro supported workload manager (WLM) to run an application on an HPE Cray EX series system
    • Use HPE Cray Supercomputing compilers to build and optimize, Fortran, C, C++, or UPC applications
    • Build and launch a parallel application using a supported version of MPI
    • Set up an interactive debugging session of a parallel application on an HPE Cray EX series system
    • Use the comparative debugger within the HPE Cray Supercomputing programming environment
    • Use the performance analysis and optimization tools to identify inefficiencies within their applications
    Divider

    Course outline

    Module 1: HPE Cray EX System Architecture and Hardware Overview

    • Login, management, scheduler, and virtualization node types
    • Liquid cooled cabinet and blade types
    • CPU types and cooling
    • GPU architecture and support
    • HPC networking, HDR Infiniband, and HPE Slingshot

    Module 2: Lustre Basics

    • Storage components and Lustre
    • Lustre node interaction
    • LFS utility, commands, and client view
    • PFL commands, use cases, and object mapping

    Module 3: HPE Cray Supercomputing Programming Environment

    • Overview, components, and versions
    • Compiler driver scripts with examples

    Module 4: PBS Pro Overview

    • PBS Pro concepts and terminology
    • PBS Pro commands, batch jobs, and scripts
    • Parallel application launch service (PALS) commands

    Module 5: Compilers in the HPE Cray Supercomputing Programming Environment

    • Overview and architecture
    • C, C++, Fortran, and Clang
    • Compiling and LoopMark listing
    • Compiler optimization: unrolling loops, vectorization, and inlining
    • Comparison of compilers used on HPE Cray Supercomputing systems

    Module 6: HPE Cray Supercomputing Programming Models and MPI

    • OpenMP, OpenACC, and MPI
    • Communication types and process
    • LIBFABRIC environment variables
    • mpich environment variables
    • MPI and OpenMP for CPU and GPU support

    Module 7: Scientific Libraries Included with HPE Cray Supercomputing Programming Environment

    • Scientific and maths libraries
    • Multi-threaded support and third-party libraries

    Module 8: Single Node Optimization

    • Low noise mode and CPU vectorization
    • Cache optimization and blocking
    • GPU optimization and programming

    Module 9: Performance Analysis and Optimization Tools

    • PerfTools and PerfTools Lite
    • PAT build, APA, and performance API
    • Hardware performance counters
    • Apprentice dashboard and sample visualizations

    Module 10: HPE Cray Supercomputing Debugging Support Tools (CDST)

    • Stack trace analysis tool (STAT) overview
    • Abnormal termination processing (ATP)
    • Gdb4hpc overview and debug sessions

    5 reasons to choose HPE as your training partner

    1. Learn HPE and in-demand IT industry technologies from expert instructors.
    2. Build career-advancing power skills.
    3. Enjoy personalized learning journeys aligned to your company’s needs.
    4. Choose how you learn: in-person, virtually, or online—anytime, anywhere.
    5. Sharpen your skills with access to real environments in virtual labs.

    Explore our simplified purchase options, including HPE Education Services – Learning Credits.

    Recommended for you