Name: Meifeng Lin
Pronouns: she/her/hers
Biography:
Dr. Meifeng Lin is a computational scientist and the group leader of the High Performance Computing group at the Computational Science Initiative of Brookhaven National Laboratory (BNL). She holds a PhD in theoretical particle physics from Columbia University, and has worked in the field of computational physics and high performance computing for more than 15 years. Prior to joining BNL in 2013, she had held postdoctoral positions at Massachusetts Institute of Technology (2007-2009), Yale University (2009-2012) and Boston University (2012-2013). She briefly held an assistant computational scientist position at Argonne National Laboratory’s Leadership Computing Facility (ALCF) before moving to BNL. Her research areas include theoretical and experimental nuclear and high energy physics, high performance computing and applications of machine learning and quantum computing in physical sciences.
Institution/Lab: Brookhaven National Laboratory
Website: www.bnl.gov/staff/mlin
SRP Collaboration Topic/Title: Exascale Computing for High Energy Physics – Addressing the Portability Challenge
Field or research area: High Performance Computing, Physics
Please select all the topical areas that apply to your project:
Computational Science Applications (i.e., bioscience, cosmology, chemistry, environmental science, nanotechnology, climate, etc.); Computer Science (i.e., architectures, compilers/languages, networks, workflow/edge, experiment automation, containers, neuromorphic computing, programming models, operating systems, sustainable software); High-Performance Computing
Brief Abstract:
Computing plays an essential role in High Energy Physics (HEP) research. From the ab initio theory calculations using numerical techniques such as lattice QCD, to detector simulations and data analysis for the particle physics experiments such as those at the Large Hadron Collider, the computational demand has been growing rapidly due to the increased, exabyte-scale, data volume from the experiments and enhanced precision needed to probe new physics beyond the Standard Model. The exascale massively parallel supercomputers hold the promise to provide orders of magnitude more computing power to support this fundamental physics research. However, the diverse heterogenous architectures in the exascale systems present many challenges to the scientific software developers. This SRP project will develop and benchmark mini code examples based on the use cases we have extracted as part of the High Energy Physics Center for Computational Excellence project (www.anl.gov/hep-cce). The goal is to use these “”mini-apps”” to track the progress of the software stack to support performance portability, the notion that a single source code can run efficiently on multiple architectures, and the ways to address it. The participants will be exposed to both the physics motifs and computational challenges in high energy physics.
Desired relevant skills, background, or interests:
Computer Science C++ Programming High Performance Computing Physics
Other comments:
The project can be tailored depending on the participant’s experience and interests. And we fully expect the SRP participants to be actively engaged with the HEP-CCE team.
Do any special requirements apply? Permanent Resident OK; International OK
Other, specify:
Keywords:
HPC; exascale; high energy physics; performance portability; computational science;
Lightning Talk Title: Big Computers Meet Big Sciences