Massachusetts Institute of Technology Postdoctoral Associate in Cambridge, Massachusetts
Job Number: 22446
Functional Area: Research - Scientific
Department: Comp Sci & Artificial Intelligence Lab
School Area: Schwarzman College of Computing
Employment Type: Full-time (Hybrid)
Employment Category: Exempt
Visa Sponsorship Available: No
Email a Friend Save Save Apply Now
Information on MIT’s COVID-19 vaccination requirement can be found at the bottom of this posting.
POSTDOCTORAL ASSOCIATE, Computer Science and Artificial Intelligence Laboratory (CSAIL), to join a project investigating the components of computing’s progress through cycle-level simulation. The past decades have seen enormous improvement in computer systems’ performance from all levels of the computing stack, including improved transistor devices, computer architecture, and algorithms. This project will investigate the relative impact of these components and aim to answer the question: how much improvement has come from improved algorithms, improvements to devices, and the computer architecture--the focus of this project. Using cycle-level simulation, the group will control for algorithmic improvement and device improvement to deeply investigate the architecture component. Will first validate that cycle-level simulation can accurately model historic computing systems; develop and parameterize open-source models in the gem5 (https://www.gem5.org/) simulator system; select and prepare a suite of workloads to use for testing and a set of historical computing systems for validation. Will then select a set of representative algorithms used across a range of domains and time periods to investigate the performance improvements of these algorithms on different computing systems and run these algorithms on top of the validated models to discover the impact of architectural improvements. Will be advised by Professor Jason Lowe-Power (https://arch.cs.ucdavis.edu/people/jason-lowe-power) of UC Davis and Dr. Neil Thompson (https://www.csail.mit.edu/person/neil-thompson) of MIT and have an opportunity to mentor undergraduate/masters/Ph.D. students at UC Davis. Full description available here (https://arch.cs.ucdavis.edu/projects/investigating-performance) .
REQUIRED: Ph.D. in computer science or related field, ideally with a background including one or more subfields of computer architecture/computer engineering; experience with computer architecture simulation; English language proficiency; excellent software engineering and communication skills; and self-motivation. PREFERRED: experience with the gem5 simulator and prior contributions to open-source software. Job #22446This is a collaboration between Dr. Neil Thompson, director of the MIT FutureTech Research Project, and Professor Jason Lowe-Power, lead PI of UC Davis Computer Architecture Research Group (DArchR). This position will involve appointments at both MIT and UC Davis. Work location may be MIT or UC Davis. 3/1/23