Required

Software Development Practices and Strategies for Efficient I/O at High Core Count of Physics Bas...

DESCRIPTION: The SDSC team presents I/O analysis using profiling/tracing tools; the focus is on understanding and improving I/O performance of scientific applications on HPC resources. PRESENTORS: Dr. Amit Majumdar, Dr. Pietro Cicotti, Dr. Mahidhar Tatineni, Dr. Manu Shantharam (SDSC, University of California San Diego) Dr. Jason Cope (DataDirect Networks), Alan Minga (Cray), Dr Sean Ziegeler (Engility, Inc.) __ DATE/LOCATION: September 29-30, 2016, Tec^Edge, Dayton OH. DISTRIBUTION STATEMENT...
Software Development...

2013 - Improving Application Performance Using TAU

DESCRIPTION: This presentation will introduce TAU's profiling, tracing, and debugging support with focus on performance data collection, analysis, and program performance optimization. PRESENTER: John Linford, Paratools, Inc. DATE/LOCATION:__July 16, 2013, ERDC DISTRIBUTION STATEMENT D. Distribution limited to the Department of Defense and U.S. DoD contractors only, for administrative or operational use, 16 July 2013. Other requests for this document shall be referred to the High Performance ...
2013 - Improving Application...

Profiling DoD HPCMP Simulation Codes Using Multiple Profilers on Multiple DSRC Systems

DESCRIPTION: This half-day seminar provides an in-depth look at Tools and Codes used for Profiling DoD Simulation Codes on a multitude of DSRC Systems, based on research conducted by Dr. Ghosh as part of a 'Tools Inter-comparison Project'. PRESENTER: Koushik Ghosh, PhD, Lawrence Livermore National Laboratory (LLNL) DATE/LOCATION: December 17, 2014, ERDC DISTRIBUTION STATEMENT D. Distribution limited to the Department of Defense and U.S. DoD contractors only, for administrative or operational ...
Profiling DoD HPCMP...

Design and Implementation of High Performance and Portable Parallel I/O and I/O Technologies for ...

DESCRIPTION: In this workshop we will present a description of the IO Interface Library (IOLib) features including its implementation details. We will introduce the IOAnalyzer tool which enables users to interactively explore and visualize profiling data generated by the Darshan tool. We will give an overview of the Scalable Checkpoint/Restart (SCR) library and how it can be used via IOLib. Finally we will present parallel I/O implementation strategies in Mini-apps. (PP-ACE-KY09-021-P3 D10) P...
Design and Implementation...

Analyzing GPU-accelerated Applications in C, C++, and Fortran with the Nvidia HPC SDK (Nov 2021)

DESCRIPTION: This training is a continuation of Developing GPU-accelerated applications in C, C++, and Fortran with the NVIDIA HPC SDK. This training will cover the basics of debugging and profiling GPU-accelerated applications using the tools in the Nvidia HPC SDK. PRESENTER(S): Ryan Montesi, GDIT/PET LOCATION: Webcast DATE & TIME: November 9, 2021, 2:00p - 3:30p ET DISTRIBUTION STATEMENT D. Distribution authorized to the Department of Defense and U.S. DoD contractors only, Administrative or...
Analyzing GPU-accelerated applications in C, C++, and Fortran with the Nvidia HPC SDK

Introduction to CUDA C/C++ (April 2022)

DESCRIPTION: In this session, NVIDIA will teach the basics of CUDA in C/C++, including basic syntax and parallel programming concepts and performance analysis on GPUs (including use of the NVIDIA Nsight profiling tools). Participants are not assumed to have any prior experience with CUDA, however experience in C/C++ is assumed. Attendees should expect to be able to learn enough to be able to port a serial CPU code to run in parallel on GPUs. The content will be introduced through a mixture of...
Introduction to CUDA C/C++

Tools and Best Practices for Developing on AMD CPUs and GPUs (May 2022)

DESCRIPTION: AMD will review product hardware architectures and highlight implications to be aware of when writing code. The lecture will be followed by demonstrations. Discussion will include: * Available toolchains, best practices, and pros and cons. * ROCm toolset, HIP, and code conversion tools. * Profiling and monitoring tools. PRESENTER(S): AMD LOCATION: Webcast DATE & TIME: May 12, 2022, 12:00p - 3:00p ET DISTRIBUTION STATEMENT D. Distribution authorized to the Department of Defense an...
Tools and Best Practices for Developing on AMD CPUs and GPUs

GPU Memory Management: Part 1 (Jun 2022)

DESCRIPTION: This training discusses GPU memory management through compiling optimization, CUDA unified memory and advanced memory management technology (prefetch, memory hints, oversubscription etc.). In this seminar we will focus on memory serial operation, blocking, stream overlap technology, and on how to avoid excess communication between CPU and GPU, and the latest approaches such as prefetching and memory advise technique. Sample codes will be adopted to demonstrate global memory and s...
GPU Memory Management: Part 1

Profiling and Analyzing HPC Applications using the HPCToolkit (Jan 2023)

DESCRIPTION: This seminar discusses the HPCToolkit, a popular tool for analyzing performance on large-scale HPC systems. This seminar includes guidance on installing the tool on DoD HPC systems, generating application profiles, and analyzing performance results. Guidance on analyzing the performance of applications on GPUs or that use OpenMP is also included. PRESENTER(S): Dr. Michael Lasinski and Paul Eller GDIT/PET LOCATION: Webcast DATE & TIME: January 17, 2023, 2:00p - 3:30p ET CONTROLLED...
Profiling and Analyzing HPC Applications using the HPCToolkit

GPU Memory Management: Part 2 Micro-benchmarks (Jun 2023)

DESCRIPTION: This seminar will discuss the TAU Performance System, a popular tool for profiling and analyzing application performance on HPC systems. This seminar will include guidance on installing this tool on HPCMP HPC systems, profiling application performance using multiple techniques, and visualizing and analyzing the results. This will include exploring multiple tools and features provided by TAU as well as highlighting the most commonly used approaches for using TAU to analyze applica...
GPU Memory Management: Part 2 Micro-benchmarks