Cristinel Ababei
Marquette University, WI·Computer Science
Scientific fields from astronomy to geoscience now produce data sets so large that simply moving the data from storage to a computer's processor becomes a bottleneck. Computational storage devices offer a different approach: they place a configurable processor, a field programmable gate array (FPGA), directly on the storage drive, so some calculations can run where the data already sits.
- Fields
Computer Science
Engineering
- Topics
computational storage
SmartSSD
field programmable gate array
- Grants
- 1 known grant, through Jun 2027
What they work on
- Explore the design space of systems pairing multi-core CPUs with SmartSSD computational storage devices to identify the computational kernels and parameters that give the best acceleration.
- Design and implement indexing data structures with a range of properties, optimized to run on FPGA-equipped computational storage devices.
- Build and share a programming framework providing data-parallel primitives (map, reduce, filter, join) and collective communication (scatter, gather, all-to-all) for computational storage.
- Implement range queries, similarity searches and joins on point and polygon data using the framework, targeting astronomy, heliophysics and geospatial applications.
More about the research
This saves energy, reduces traffic across the system's limited connections, and cuts waiting time.
Why it matters
Researchers in astronomy, heliophysics, geoscience and other geospatial fields increasingly struggle to search and compare huge data sets quickly. By moving computation onto storage devices, this work could make these searches faster and more energy efficient. A shared programming framework could let scientists use the hardware without specialized expertise.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Satish Puri
Missouri University of Science and Technology, MO·Computer Science·Early-career
Find which computing tasks and design settings gain the most speed on SmartSSD computational storage devices, which pair a drive with an FPGA chip, through a design space exploration on multicore systems.
Bo Yuan
Rutgers University-New Brunswick, NJ·Engineering·Early-career
Develop design principles for building domain expertise into learning-based robot motion planners, improving their reliability and performance in complex, resource-limited real-world settings.
Mahmut Kandemir
Pennsylvania State University, PA·Computer Science
Characterize diverse HPC workloads on real and simulated computational storage devices to identify which computations benefit from running inside storage, and what performance tradeoffs result.
The school
Marquette University: 4 labs on PI Pages, $49M a year on research, 69 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.