A team led by Princeton Plasma Physics Laboratory’s C.S. Chang recently used the 27-petaflop Titan supercomputer to simulate a crucial transition phenomenon in MIT’s Alcator C-Mod tokamak, uncovering the basic physics behind the transition.
In June, the OLCF’s Tom Papatheodore led a CUDA workshop to teach students, interns, and researchers how to program on a GPU architecture such as Titan.
OLCF staff members recently participated in coding events that aimed to introduce middle and high school students to computing and coding concepts—CodeStock Academy and WiC’s “Introduce Your Daughter to Code.”
A team led by University of Iowa’s George Constantinescu is using Titan to create 3-D non-hydrostatic flood models that can be used to improve the predictive capabilities of existing 2-D models.
The 2017 OLCF user meeting, held May 23–25 at ORNL, gave users and staff a chance to share achievements on Titan, discuss Summit, and explore deep learning.
A team led by ORNL’s Amit Shyam and Dongwon Shin is using Titan to explore the possibilities of designing various high-temperature–capable alloys, in hopes of changing the paradigm for current alloy design and significantly shortening the typical alloy development and deployment process.
The OLCF’s new NVIDIA DGX-1 deep learning system is offering scientists opportunities to explore deep learning’s potential to leverage big data analytics to automate and accelerate the scientific discovery process.
The OLCF played a major role in the annual American Physical Society March Meeting—the largest gathering of physicists in the world—by bringing high-performance computing talks to the meeting as part of a petascale computing focus session.
OLCF staff members recently built and ran a technology called containers, which bundle an operating system and software into a single file and make it easier for researchers to run deep learning software on OLCF supercomputers.
The OLCF’s new ARM1 early development system gives researchers the opportunity to test various software packages and explore an experimental environment for ARM architecture–based systems.