Exascale Computing Postdoctoral Fellow
- Lawrence Berkeley National Laboratory
- Location: Berkeley, CA
- Job Number: 7086667 (Ref #: 33751)
- Posting Date: Oct 6, 2021
- Application Deadline: Open Until Filled
Job DescriptionExascale Computing Postdoctoral Fellow - 93950
Lawrence Berkeley National Lab’s (LBNL, https://www.lbl.gov/) NERSC Division (https://www.nersc.gov/about/) has an opening for an Exascale Computing Postdoctoral Fellow to join the team.
The National Energy Research Scientific Computing Center (NERSC) at Lawrence Berkeley National Laboratory seeks a highly motivated postdoc to join the NERSC efforts to ensure math libraries perform well on upcoming computational platforms. The postdoc will collaborate with selected Exascale Computing Project (ECP) math libraries and scientific application teams to enable the solution of deep, meaningful problems targeted by the ECP program as well as other program areas funded by the Department of Energy Office of Science (https://www.energy.gov/science/office-science).
NERSC provides world-class supercomputing, high-performance, scalable data systems and services to 7000 users across more than 700 projects running 800 different codes. NERSC’s science impact is acknowledged by over 2000 publications per year.
The Challenge: Enabling advanced simulation and modeling at scale on energy-efficient supercomputers. In 2020, NERSC began deploying its first production heterogeneous CPU/GPU-based HPE/Cray supercomputer, “Perlmutter.” Perlmutter, a system optimized for science, includes the latest AMD CPUs and NVIDIA GPUs, a high-speed interconnect, and an all-flash file system. DOE will also be deploying the following two exascale systems. The Oak Ridge Leadership Computing Facility (OLCF) will be deploying the Frontier system with the same high-speed interconnect with AMD CPUs and GPUs. The Argonne Leadership Computing Facility (ALCF) will be deploying the Aurora system with the same high-speed interconnect with Intel CPUs and GPUs. Many applications running at NERSC as well as OLCF and ALCF must be adapted or optimized to run efficiently on GPUs, and solutions that put GPU performance in the hands of users must be portable ones.
Cutting-edge simulation of complex physical phenomena after the discretization of the governing physical equations often boils down to a numerical solution, and performance of solver and math libraries is critical to the performance of the application. It has been challenging to implement many of these solvers/math libraries on GPUs and to ensure that they perform well on GPUs from three vendors.
As a NERSC solvers/math libraries postdoctoral fellow you will work with one or more of the ECP math libraries teams to transition and optimize codes for the Perlmutter system and the first two DOE exascale machines and produce mission-relevant science that pushes the limits of HPC. An important aspect is to ensure the math libraries perform well in the applications. ECP math libraries of interest include PETSc, SLATE, Ginkgo, SuperLU, etc.
What You Will Do:
• Work with NERSC staff and code teams to transition and optimize math libraries for ECP applications and selected non-ECP applications important to the NERSC workload for the Perlmutter system as well as the upcoming OLCF Frontier and ALCF Aurora systems in performance-portable ways.
• Conduct profiling and scaling studies as well as parallelization, memory bandwidth for these codes.
• Disseminate results of research activities through refereed publications, reports, and conference presentations. Ensure that new methods are documented for the broader community, NERSC staff, vendors, and NERSC users.
• Participation in postdoctoral career and science enrichment activities within the Berkeley Lab Computing Sciences Area is encouraged.
• Occasional to travel to sites at other labs, universities, and to vendor facilities.
What is Required:
• Ph.D. in applied mathematics, computational science, computer science, or a science domain area with a computationally-oriented research focus including but not limited to mathematics, physics, engineering, etc. Expertise in applied mathematics is preferred.
• Research experience and knowledge in computing and/or code development for HPC, algorithm design, or applied mathematics.
• Effective communication and interpersonal skills.
• Ability to work productively both independently and as part of an interdisciplinary team balancing objectives involving research and code development.
• Experience with GPU and parallel/manycore computer architectures, threading, and vectorization.
• Experience with the development and performance optimization of scientific software in the HPC context.
• Publication record or contributions to open-source software projects commensurate with years of experience.
• Experience with numerical linear algebra, particle methods, or grid methods.
• Experience with C, C++, Fortran, MPI, threading, or data structure transformations.
• This is a full-time 2-year postdoctoral appointment with the possibility of renewal based upon satisfactory job performance, continuing availability of funds, and ongoing operational needs. You must have less than 3 years of paid postdoctoral experience. Salary for Postdoctoral positions depends on years of experience post-degree.
• This position is represented by a union for collective bargaining purposes.
• Salary will be predetermined based on postdoctoral step rates.
• This position may be subject to a background check. Any convictions will be evaluated to determine if they directly relate to the responsibilities and requirements of the position. Having a conviction history will not automatically disqualify an applicant from being considered for employment.
• Work will be primarily performed at Lawrence Berkeley National Lab, 1 Cyclotron Road, Berkeley, CA.
How To Apply
Apply directly online at http://22.214.171.124/counter.php?id=209415 and follow the on-line instructions to complete the application process.
Berkeley Lab is committed to Inclusion, Diversity, Equity, and Accountability (IDEA, https://diversity.lbl.gov/ideaberkeleylab/) and strives to continue building community with these shared values and commitments.
Berkeley Lab is an Equal Opportunity and Affirmative Action Employer. We heartily welcome applications from women, minorities, veterans, and all who would contribute to the Lab’s mission of leading scientific discovery, inclusion, and professionalism. In support of our diverse global community, all qualified applicants will be considered for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, age, or protected veteran status.
Equal Opportunity and IDEA Information Links:
Know your rights, click here for the supplement: "Equal Employment Opportunity is the Law" and the Pay Transparency Nondiscrimination Provision (https://www.dol.gov/sites/dolgov/files/ofccp/pdf/pay-transp_%20English_formattedESQA508c.pdf) under 41 CFR 60-1.4.