NERSCPowering Scientific Discovery Since 1974

GPUs for Science 2020

Location: Zoom Webinar

June 30 - July 1, 2020

In anticipation of our next system Perlmutter which will arrive soon, NERSC has a crucial role to play in helping the wider scientific community leverage and embrace GPUs for their simulation, data processing, and machine learning workloads. Part of this role includes hosting events like GPUs for Science, the goal of which is to help facilitate the transition to GPU systems by giving scientists the motivation, the tools, and the expertise they need to make this change possible. 


Day 1 Agenda: Tuesday, June 30

Registration   8:45 AM
Welcome   9:00 AM
Perlmutter Introduction Jay Srinivasan (LBNL) 9:20 AM
TALK: Using GPUs as accelerators Max Katz (NVIDIA) 9:40 AM
TALK: Using GPUs as accelerators Rene Van-Oostrum (AMD) 10:00 AM
TALK: Using GPUs as accelerators Jeff Hammond (INTEL) 10:20 AM
Coffee Break (20 mins) Zoom Breakout Rooms 10:40 AM
PANEL: Performance & Portability Christopher Daley (LBNL), John Owens (UC Davis), Phil Roth (Oak Ridge National Lab) 11:00 AM
TALK: Mixed precision optimization for lattice quantum chromodynamics Kate Clark (NVIDIA) 12:00 NOON
TALK: New protein crystallographic science enabled with GPU computing at X-ray free electron lasers Nicholas Sauter (LBNL) 12:20 PM
Lunch Break (1 hour) Zoom Breakout Rooms 12:40 PM
TALK: Science applications: ExaFEL project port to GPU Johannes Blaschke (LBNL) 1:40 PM
TALK: Science applications: Using OpenACC to accelerate scientific applications on GPUs Sunita Chandrasekaran (University of Delaware) 2:00 PM
TALK: Science applications: Application of GPUs in Proteomics and Connectomics Fahad Saeed (FIU) 2:20 PM
TALK: Science applications: GPU-Accelerated General Relativistic Ray Tracing for Simulating Black Hole Images Chi-kwan Chan (Steward Observatory/University of Arizona) 2:40 PM
Coffee Break (20 mins) Zoom Breakout Rooms 3:00 PM
TUTORIAL: OpenACC Programming Jeff Larkin (NVIDIA) & Rahul Gayatri (LBNL) 3:20 PM
End of Day 1   5:00 PM


Day 2 Agenda: Wednesday, July 1

Registration   8:45 AM
TALK: Block Copolymer Assembly using Reinforcement Learning on GPUs Thorsten Kurth (NVIDIA) 9:00 AM
TALK: Experience in Porting Science Applications to GPUs Charlene Yang (LBNL) 9:20 AM
PANEL: Programming platforms: Kokkos, Raja and OpenACC Rahul Gayatri (LBNL), Sunita Chandrasekaran (University of Delaware), David Alexander Beckingsale (LLNL)
9:40 AM
Coffee Break (20 mins) Zoom Breakout Rooms 10:40 AM
TALK: Introduction to the Roofline Model Samuel Williams (LBNL) 11:00 AM
TALK: Profiling/debugging Jonathan Madsen (LBNL) 11:20 AM
PANEL: Reduced/Mixed Precision Optimization Techniques CJ Newburn (NVIDIA), Xiaoye Sherry Li (LBNL) & Cindy Rubio González (UC Davis) 11:40 AM
Lunch Break (1 hour) Zoom Breakout Rooms 12:40 PM
TALK: Science applications: DESI Laurie Stephey (LBNL) 1:40 PM
TALK: Science applications: DL4Science Steven Farrell (LBNL) 2:00 PM
TUTORIAL: Intro to CUDA programming (Part 1) Kevin Gott & Jonathan Madsen (LBNL) 2:20 PM
Coffee Break (20 mins) Zoom Breakout Rooms 3:00 PM
TUTORIAL: Intro to CUDA programming (Part 2) Kevin Gott & Jonathan Madsen (LBNL) 3:20 PM
Lessons Learned: Summary of the event Organising Committee 4:20 PM
Closing Remarks and Thanks Jack Deslippe (LBNL) 4:40 PM
End of Day 2   5:00 PM



Registration is FREE and open for everyone interested, and users need to sign up at this link. There are limited spaces available in the tutorial sessions, and attendees are encouraged to sign up as early as possible to ensure they get their preferred tutorial session(s).

Registration deadlines:

  • General attendance: Sunday, June 28th, 11:59PM (PDT).
  • Tutorial sessions: Wednesday, June 23rd, 11:59PM (PDT).

We will use for communication during the event, you can contact one of our organizers in case you need access. Once registered please join the GPU for Science (gpu4science) channel.



The members of the organizing committee for GPUs for Science 2020 are (in alphabetical order by last name):

Muaaz Awan
Hugo Brunie
Oisín Creaner
Dossay Oryspayev
Michael Rowan
Yan Zhang



Zoom Web conference: The Zoom link will be published shortly before the event. 

About LBNL

Lawrence Berkeley National Laboratory is a multidisciplinary national laboratory located in Berkeley, California on a hillside directly above the campus of the University of California at Berkeley. The site consists of 76 buildings located on 183 acres, which overlook both the campus and the San Francisco Bay. »More Visitor Information.