Job Description
The Mathematics and Computer Science Division at Argonne National Laboratory seeks well-prepared candidates for a postdoctoral position in parallel discrete event simulations (PDES) of HPC networks and workloads.
In this role you will investigate new approaches to combining PDES with surrogate modeling techniques, such as machine learning and statistical models, to accelerate hardware co-design simulations of large-scale supercomputers. Models of HPC networks and communication-intensive workloads are expected to be key elements of this work.
You will also actively collaborate with domain scientists, university researchers, and supercomputer vendors and have the opportunity to build an independent research program. The expected work will provide the opportunity to use advanced supercomputers across the Department of Energy computing facilities, such as the planned Aurora exascale system (https://www.alcf.anl.gov/aurora) at the Argonne Leadership Computing Facility, to develop and evaluate simulation models at scale.
Required skills and qualifications:
PhD (recent or soon-to-be completed within the last 0-3 years) in computer science or a related discipline.
Graduate/postgraduate research experience in computer science and/or high-performance computing.
Programming experience in C/C++ and Python.
Self-motivated with the ability to work effectively both independently and collaboratively
Ability to model Argonne's core values: impact, safety, respect, integrity, and teamwork.
Preferred knowledge (not required):
Research or development experience with PDES
Experience using PDES toolkits such as CODES/ROSS
(https://github.com/codes-org/codes/wiki) and SST (http://sst-simulator.org/)
US citizenship is NOT required for this position. |