HPC Application & Workflow Positions

Princeton University's Cooperative Institute for Modeling the Earth System (CIMES) has openings in the general area of high-performance computing, algorithms and workflows. The positions will support an exciting new initiative to adapt GFDL/CICS Earth System models to novel architectures and systems.

Our effort will take place in the context of an integrated effort to target NOAA Earth system models to these novel architectures, where we will contribute to the overall NOAA effort, as well as specifically target models and model components used in GFDL Earth System Modeling and based on the Flexible Modeling System (FMS). (http://www.gfdl.noaa.gov/fms).

The scope of the work is broad. It includes:

  • redesigning the architecture of the Earth system model for greater concurrency
  • the optimization of codes for fine-grained parallel architecture, requiring knowledge of compilers (including layers like LLVM) and languages for expressing parallelism (MPI, OpenMP, OpenACC, CUDA, PGAS)
  • the redesign and substitution of existing algorithms with others better suited for novel architectures.
  • the redesign of the complete workflow systems for maximizing system throughput.

Candidates will join a dynamic team of computational experts attached to one of the foremost research institutions in Earth system science, will collaborate with top scientists and algorithm developers in the field, and will have access to an extraordinary range of computational and data resources. This is a growth area for the lab and the lab's group leading this effort, and candidates will have considerable leeway to develop their own ideas within the context of the lab's mission. Candidates must be able to work in a team environment that combines the collegiality of an academic setting with the focus of a mission agency, and be able to deliver rigorously tested results to meet deliverables for international collaborative science missions. Occasional travel to project meetings, site visits, national and international conferences, and workshops is to be expected.

Candidates must possess an MS in a relevant field (computer science or physical sciences) and at least two years experience working with weather or climate codes and runtime environments in an HPC context. Knowledge of one or more parallel programming methods is desirable. Appointments will be made at either the Associate Professional Specialist or Professional Specialist rank, depending on the candidate's credentials. At least one senior hire (PhD and/or 10 years relevant experience) is envisioned.

Curriculum vitae, a publication list, and contact information for three references, should be submitted online to https://www.princeton.edu/acad-positions/position/11883.

Princeton University is an Equal Opportunity/Affirmative Action employer and all qualified applicants will receive consideration for employment without regard to age, race, color, religion, sex, sexual orientation, gender identity or expression, national origin, disability status, protected veteran status, or any other characteristic protected by law. This position is subject to the University's background check policy.

posted: 22 May 2019     Please mention EARTHWORKS when responding to this advertisement.