CUDA Multithreading with Streams, July 16, 2021

Introduction

NVIDIA will present “CUDA Multithreading with Streams” to OLCF and NERSC users on Friday, July 16, 2021. This event is a continuation of the CUDA Training Series.

CUDA Streams are a useful way to achieve concurrency and ensure that an application is fully utilizing the GPU. However, many GPU-accelerated applications start off as multi-threaded CPU applications. In this module, we will explore the combination of GPU concurrency and a multi-threaded context, such as an OpenMP application.

After the presentation, there will be a hands-on session where participants can complete example exercises meant to reinforce the presented concepts and observe their impact on application performance.  Cori-GPU access will be provided for current NERSC users for the hands-on exercises.  Temporary OLCF Summit access will not be available for remote participants.

Date and Time: 10 am - 12 pm (Pacific time), Friday, July 16, 2021

The format of this event will be online only. 

Registration

Registration is required for remote participation.  Please click the "Registration" drop down on this page to register.  

Remote Connection Information 

Registration is required for remote participation.  Please click the "Remote Connection Details" drop down on this page to register. 

• Slides
• Recording
• Exercises: The example exercises for this module can be found in the "exercises/hw10" folder of this GitHub repo.