Near-memory primitive support and infratructure for sparse algorithm

Author(s)
Garg, Kartikay
Advisor(s)
Yalamanchili, Sudhakar
Young, Jeffrey
Editor(s)
Associated Organization(s)
Supplementary to:
Abstract
This thesis introduces an approach to solving the problem of memory latency performance penalties with traditional accelerators. By introducing simple near-data-processing (NDP) accelerators for primitives such as SpMV (Sparse Matrix Multiplication of Vectors) and DGEMM (Double Precision Dense Matrix Multiplication) kernels, applications can achieve a considerable performance boost. We evaluate our work for SuperLU application for the HPC community. Thesis Statement: Reevaluating core primitives such as DGEMM, SCATTER, and GATHER for 3D-stacked PIM architectures that incorporate re-configurable fabrics can deliver multi-fold performance improvements for SUPERLU and other sparse algorithms.
Sponsor
Date
2017-04-28
Extent
Resource Type
Text
Resource Subtype
Thesis
Rights Statement
Rights URI