NESAP for Data Postdoctoral Fellow
- Lawrence Berkeley National Laboratory
- Location: Berkeley, CA
- Job Number: 7070834 (Ref #: 11783)
- Posting Date: Oct 9, 2020
- Application Deadline: Open Until Filled
Job DescriptionNESAP for Data Postdoctoral Fellow - 91231
To address today’s most urgent research questions, scientists are collecting exponentially more data, managing it with automated or interactive software pipelines, and analyzing it using algorithms of ever-increasing computational intensity. In order to support these scientists, the National Energy Research Scientific Computing (NERSC, https://www.nersc.gov/) Center at Berkeley Lab (https://www.lbl.gov/) is deploying a new supercomputer in 2021 named “Perlmutter,” a system with 3-4 times the performance of NERSC’s current supercomputer “Cori.” Perlmutter, NERSC’s first GPU+CPU supercomputer, has a design optimized for science, but now NERSC faces the challenge of helping its users do their science optimally with Perlmutter. That is where you come in.
We are looking for highly motivated postdocs to join the NERSC Exascale Application Program (NESAP, https://www.nersc.gov/research-and-development/nesap/) for Data (N4D). N4D addresses data-intensive science pipelines that process massive datasets from facilities like synchrotron light sources, telescopes, microscopes, particle accelerators, or genome sequencers. The impact from accelerating the analysis of data from these instruments cannot be understated: predicting materials for new, efficient solar cells and batteries, understanding fundamental biological processes, unlocking the secrets of the Universe. Our vision is seamless integration and flow of data between scientific facilities and energy-efficient supercomputers to enable discoveries from big team science. Learn more about the kind of science supported by N4D here (https://www.nersc.gov/research-and-development/nesap/nesap-projects/#n4d).
As a NESAP postdoc you will:
• Collaborate with an experimental or observational data science team to ensure that their code, workflows, and software infrastructure make the best use of Perlmutter possible.
• Be part of a multidisciplinary team composed of computational and domain scientists working together to transition and optimize codes to the Perlmutter system and produce mission-relevant science that pushes the limits of high-performance computing (HPC).
• Carry out code transition efforts in collaboration with a project’s PI and team members, supported by NERSC and vendor staff.
• Collaborate with and support other postdocs across all three NESAP program areas (simulation, data, and learning). Learn about current and previous NESAP postdocs here.
• Have the opportunity to push the boundaries of science by assisting in the development of new methods and implementation of novel algorithms on state-of-the-art computational resources.
NERSC’s Perlmutter system has both a GPU and CPU partition. The GPU partition has 1500 nodes each with 4 NVIDIA A100 (Ampere) Tensor Core GPUs, one AMD Milan CPU, and 256 gigabytes of memory. The CPU partition has roughly 3000 CPU nodes each with 2 AMD Milan CPUs and 512 gigabytes of memory. Perlmutter also features a new Cray system interconnect codenamed Slingshot that is designed for data-centric computing, and a new 35-petabyte all-flash Lustre scratch file system designed to move data at a rate of more than 5 terabytes per second. More details on Perlmutter are available here (https://www.nersc.gov/systems/perlmutter/).
Successful Candidates are expected to:
• Work with NERSC staff, code teams, and HPC vendor partners to transition and optimize codes, pipelines, and workflows for Perlmutter that process or analyze data from scientific facilities, balancing performance, productivity, and portability.
• Conduct profiling, scaling, parallelization (GPU and/or CPU), memory bandwidth, and I/O performance analyses to guide development of these codes, pipelines, and workflows; capitalize on NERSC’s combined HPC and Data ecosystems.
• Disseminate results of research activities through refereed publications, reports, and conference presentations. Ensure that new findings are documented for the broader community, NERSC staff, vendors, and NERSC users.
• Participate in postdoctoral career development, science enrichment, and networking opportunities within the Computing Sciences Area and broader Berkeley Lab community.
• Travel to sites at other Labs, universities, and vendor facilities in order to collaborate with teams, experts, and vendor staff.
What is Required:
• Ph.D. in Computational Science, Data Science, Computer Science, Applied Mathematics, or a science domain area with a computationally-oriented research focus.
• Experience, knowledge, or willingness to learn about code, pipeline, or workflow development for experimental/observational science and HPC.
• Ability to work productively either independently or as part of an interdisciplinary team, balancing objectives involving both research and development.
• Effective communication and interpersonal skills.
• Experience with development and performance optimization of scientific software using modern best practices (revision control with Git, continuous integration and deployment, etc).
• Experience with at least one high-level language (HLL) such as Python, Julia, or R and corresponding data analytics ecosystem. Bonus: experience with CPU/GPU parallelism.
• Interest in one or more of the following areas: container technologies (e.g. Docker), Jupyter notebooks, complex workflows and pipelines, data analytics in HPC.
How To Apply
Apply directly online at http://220.127.116.11/counter.php?id=187447 and follow the on-line instructions to complete the application process.
To be considered applications must include:
• A Cover Letter: Include a cover letter introducing yourself, your application materials, and describing your interest in the program. How did you hear about NESAP for Data and why do you want to be a part of it?
• Curriculum Vitae/Resume: Either an academic CV or a resume is acceptable. Please be sure to include these particularly relevant facts: technical skills, interests, and research/engineering activities relevant to the position.
• List of Publications: A list of publications is required in order to show evidence that you have experience with the academic/conference publishing process. If there’s a particular paper you would like us to look at, please highlight that in the list (please don’t attach the paper itself). Links to software projects and public code repositories in your publication list are also very welcome!
• 3 References: Provide contact information for three professional references with whom we may communicate regarding your work and your application. Please make sure they agree to provide a reference for you.
• This is a full-time 1 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 4 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.
• Work will be primarily performed at Lawrence Berkeley National Lab, 1 Cyclotron Road, Berkeley, CA.
Equal Employment Opportunity: Berkeley Lab is an Equal Opportunity/Affirmative Action Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, age, or protected veteran status. Berkeley Lab is in compliance with the Pay Transparency Nondiscrimination Provision under 41 CFR 60-1.4. Click here (https://www.dol.gov/agencies/ofccp/posters) to view the poster and supplement: "Equal Employment Opportunity is the Law."
Lawrence Berkeley National Laboratory encourages applications from women, minorities, veterans, and other underrepresented groups presently considering scientific research careers.