Abstract
dc:description.abstractThe objective of the proposed research is to explore adding a sparse tensor core to GPGPUs to accelerate GEMM workloads in AI. With the increasing use of GPUs for AI training and inference, there is a growing need to add specialised hardware to GPUs to reduce the execution time for inference and training. A sparse matrix multiply unit has the potential to speed up this process significantly, and the objective of this research is to test this core in different benchmarks in a GPGPU called Vortex to test its viability.
Degree
thesis:*- Name thesis:degree_name
- Electrical and Computer Engineering, MS
- Department thesis:degree_department
- Electrical and Computer Engineering
- Grantor
- Georgia Institute of Technology
- Year dc:date.issued
- 2026
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Gandhi, Krishil Rakeshkumar
- Advisor dc:contributor.advisor
-
- Kim, Hyesoon
- Committee members dc:contributor.committeemember
-
- Mahajan, Divya
- Krishna, Tushar
Identifiers
dc:identifier.*- Handle dc:identifier.uri
- https://hdl.handle.net/1853/81613