ESP Training Series: Overview of OpenMP 4.5 and 5.0 Features

Sales Have Ended

Registrations are closed
Registration is now closed.

Event Information

Share this event

Date and Time

Sales Have Ended

Registrations are closed
Registration is now closed.
Event description

Description

The Aurora ESP Training Series


An Overview of OpenMP 4.5 and 5.0 Features


Speakers

Jose M. Monsalve Diaz, University of Delware; Colleen Bertoni and JaeHyuk Kwack, Argonne Leadership Computing Facility



Abstract

The Aurora ESP Training Series continues on June 18, with a webinar focused on OpenMP 4.5 and 5.0 features. The webinar will consist of three segments:

- A review of the fork-join OpenMP programming model, followed by an introductory overview of the tasking model. This segment will provide application developers with the necessary tools to go beyond the commonly used parallel loop.

- A brief introduction to the OpenMP execution model and data environment for accelerators such as GPUs. This segment will also discuss the important basic OpenMP 4.5 constructs and concepts.

- A review of the OpenMP standard, 5.0, and its new constructs. This segment will cover several important features in OpenMP 5.0 that can improve usability, offload flexibility, and extend the existing functionality in OpenMP 4.5.


About the Speakers

Jose M. Monsalve Diaz is a fifth-year PhD student from the University of Delaware. He has worked on parallel programming models and dataflow based parallel computer architectures for HPC. Jose is currently studying program execution models for parallelism, seeking to stablish a hardware-software interface that allows a cleaner overview of a program executing in a parallel system and has been developing an OpenMP Validation and Verification (OMPVV) test suite. This project aims to test compilers and HPC systems with respect to the OpenMP specifications.

Colleen Bertoni is an assistant computational scientist in the Performance Engineering group at the Argonne Leadership Computing Facility. Her research interests include programming models for high-performance computing and electronic structure theory.

JaeHyuk Kwack is an assistant computational scientist in the Performance Engineering group at the Argonne Leadership Computing Facility (ALCF). His interests include computational fluid dynamics (CFD) and fluid-solid interaction (FSI). Before joining ALCF, he worked as a research programmer for the Blue Waters supercomputing project at National Center for Supercomputing Applications (NCSA). Since joining Argonne in 2018, he has been working on performance tools, math libraries, and programming models for the ALCF’s future exascale system.


About the Aurora ESP Training Series

This training series is for Aurora Early Science Program project teams. We will focus on topics and tools relevant to current production machines, with an emphasis on data intensive and machine learning subjects.



Share with friends

Date and Time

Save This Event

Event Saved