Jose Martinez
Cornell University, NY·Computer Science
Computer processors have become much faster than the memory chips that feed them data. Programs that sift through huge datasets, such as the database queries businesses use to make decisions, therefore spend much of their time waiting on memory.
- Fields
Computer Science
Engineering
- Topics
processing in memory
in-memory computing
computer architecture
- Grants
- 1 known grant, through Sep 2027
What they work on
- Develop mechanisms for running database analytics queries in place inside DRAM, a computer's main memory, and explore how such intelligent DRAM can cooperate with other processing units.
- Design processing-in-memory schemes for SRAM-based caches, the fast on-chip memory, and test whether associative processing, which operates on many stored items at once, suits data analytics workloads.
- Create a domain-specific language based on an operational algebra that decomposes queries into dataflow graphs and optimizes their execution across SRAM-based and DRAM-based processing-in-memory units and the CPU.
- Build a simulation infrastructure and benchmark suite for evaluating combined database-hardware designs, usable by the broader computer architecture and database research communities.
More about the research
For decades chips got faster every generation as transistors shrank, but that steady improvement has hit physical limits, so faster analytics needs new hardware ideas. This project explores "intelligent" memory that computes results right where the data is stored, and redesigns the hardware and software of the analytics pipeline together so the two work as a pair.
Why it matters
Many industries depend on fast analysis of growing data volumes to make decisions, and conventional processors are increasingly limited by how quickly they can move data from memory. Memory that can compute could make database analytics substantially faster and cheaper to run. The planned simulators and benchmarks could also give other computer architects and database researchers a shared way to test new designs.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Kevin Skadron
University of Virginia, VA·Computer Science
Develop ways to run database analytics queries directly inside DRAM main memory, and explore how intelligent DRAM can work together with other processing units.
Josep Torrellas
University of Illinois Urbana-Champaign, IL·Computer Science
Develop approximate algorithms for graph clustering in static and dynamic settings, graph construction that preserves key information, and machine learning, for computational biology, social network and software verification problems.
Tushar Krishna
Georgia Institute of Technology, GA·Computer Science
Extend the gem5 computer simulator to model emerging devices and effects, including chiplets (small chips combined in one package), hardware reliability and security.
The school
Cornell University: 37 labs on PI Pages, $1.6B a year on research, 537 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.