Tushar Krishna
Georgia Institute of Technology, GA·Computer Science
As the steady gains in chip speed predicted by Moore's Law slow down, computers are getting faster mainly by growing larger and more specialized. Designing them well now means thinking about the whole stack at once, from transistors up to applications.
- Fields
Computer Science
Engineering
- Topics
gem5
computer architecture
computer simulation
- Grants
- 1 known grant, through Sep 2027
What they work on
- Extend the gem5 computer simulator to model emerging devices and effects, including chiplets (small chips combined in one package), hardware reliability and security.
- Improve gem5's models of modern hardware by increasing their accuracy, adding support for new vector extensions, and strengthening support for accelerators such as graphics processing units (GPUs).
- Speed up gem5 simulations and make gem5 interoperate with other simulation tools so that larger, higher-fidelity computer systems can be studied.
- Make gem5 easier to reuse and reproduce by providing simpler baseline systems, ready-to-use benchmarks, and workshops, tutorials, bootcamps and self-paced teaching tools.
More about the research
Researchers cannot build every candidate design in silicon, so they rely on simulators, software that models a computer in detail. This project expands gem5, a widely used open-source computer simulator, so that it can model more kinds of systems more accurately and serve more research communities.
Why it matters
Better simulation tools could help engineers design and understand the next generation of supercomputers, laptops and mobile devices before they are built. A faster, more accurate and easier-to-use gem5, with shared benchmarks, could make hardware research more reproducible and open the field to more students and scientific communities.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Divya Mahajan
Georgia Institute of Technology, GA·Computer Science·Early-career
Build a cross-platform framework that measures how thermal hotspots, frequency throttling, power spikes and energy limits restrict distributed AI training performance on GPUs and other accelerators.
Shivaram Venkataraman
University of Wisconsin-Madison, WI·Computer Science·Early-career
Characterize performance variability across many types of accelerator hardware, where identical applications run at different speeds on chips of the same type, and identify which applications are most affected (Fortuna).
Omiya Hassan
Boise State University, ID·Engineering
Co-design an electro-optical application-specific integrated circuit that pairs photonic (light-based) and electronic processing units, assigning AI training and inference to the best-suited unit.
The school
Georgia Institute of Technology: 50 labs on PI Pages, $1.5B a year on research, 572 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.