CUDA Concurrency

CUDA Concurrency
Tuesday, July 21, 2020

On-site participation will be held at the following sites:

ORNL Building 5100, Room 128 (JICS Auditorium) 1:00 PM – 3:00 PM (ET)
NERSC Building 59 (Shyh Wang Hall), Room 4102 10:00 AM – 12:00 PM (PT)

On Tuesday, July 21, 2020, NVIDIA will present part 7 of a 9-part CUDA Training Series titled “CUDA Concurrency”.

Concurrency refers to the ability to perform multiple CUDA operations simultaneously. This allows the user to do things like run multiple kernels on a single GPU or overlap data transfer and computation. This module introduces CUDA Streams and asynchronous functions that can be used to achieve concurrency in your applications. After the presentation, there will be a hands-on session where participants can complete example exercises meant to reinforce the presented concepts.

In-Person Participation
OLCF and NERSC will both be holding in-person events for each part of the series, where participants can watch the presentations and get help from experts during the hands-on sessions. In-person participants without current Summit or Cori-GPU access will be given temporary accounts to work on the examples.

NOTE: The deadline for in-person registration is July 13, 2020.

Remote Participation
Remote participants can watch the presentations via web broadcast and will have access to the training exercises, but temporary access to the compute systems will be limited as follows:

  • Current NERSC users will have Cori-GPU access temporarily added to their accounts.
  • Temporary Summit access will not be available for remote participants.

NOTE: Registration is required for both in-person and remote participation. To register, please submit the form below.

If you have any questions, please contact Tom Papatheodore (

Remote connection details will be added here the day before the event.

The slides will be added here before the event and the recording will be added shortly after the event.

Shortly before the event, the example exercises for this module will be added to the exercises/hw7 folder of the following GitHub repo:

CUDA Concurrency - Survey

  • Rate from 1 to 5 (with 5 being highest)
  • Rate from 1 to 5 (with 5 being highest)
  • Rate from 1 to 5 (with 5 being highest). Select N/A if you did not work on the exercises.
  • Rate from 1 to 5 (with 5 being highest). Select N/A if you did not work on the exercises or if you completed the exercises on your own.


Jul 21 2020


1:00 pm - 3:00 pm


BLDG 5100, RM 128 (JICS Auditorium)


Thomas Papatheodore
1 (865) 576-1244
QR Code

Comments are closed.