- Application Opportunities and Challenges on the Titan Supercomputer: Accelerating the Path to the Exascale
- Guest Speaker
- Jack C. Wells
- Guest Affiliation
- Director of Science, Oak Ridge Leadership Computing Facility, ORNL
- Prof. David Landau, firstname.lastname@example.org
- Thursday, February 25, 2016 4:00 pm - 5:00 pm
- Physics Auditorium (Rm 202)
Modeling and simulation with Petascale computing has supercharged the process of innovation, dramatically accelerating time-to-insight and time-to-discovery. The Titan supercomputer is the Department of Energy, Office of Science’s flagship Cray XK7 supercomputer managed by the Oak Ridge Leadership Computing Facility (OLCF). With its hybrid, accelerated architecture of traditional CPUs and graphics processing units (GPUs), Titan allows advanced scientific applications to reach speeds exceeding 20 petaflops with a marginal increase in electrical power demand over the previous generation leadership-class supercomputer. I will summarize the benefits, challenges, and lessons learned in deploying Titan and in preparing applications to move from conventional CPU architectures to a hybrid, accelerated architectures, and how the evolution of workloads on Titan is providing new opportunities for research in the areas of workflows and adaptive system schedulers. I will emphasize the challenges users have encountered with emerging programming models and how uses are addressing them using languages, libraries, and compiler-based solutions.
I also plan to discuss the science outcomes from Titan in diverse areas such as physics, materials sciences, nuclear energy, and engineering sciences. I will also discuss research outcomes from a growing number of industrial partnerships. We will discuss implications for the research community as we prepare for exascale computational science and engineering within the next decade. I will also provide an overview of user programs at the Oak Ridge Leadership Computing Facility with specific information how researchers may apply for allocations of computing resources.