research
∙
03/08/2023
Cascading GEMM: High Precision from Low Precision
This paper lays out insights and opportunities for implementing higher-p...
share
research
∙
02/16/2023
GEMMFIP: Unifying GEMM in BLIS
Matrix libraries often focus on achieving high performance for problems ...
research
∙
04/11/2019
The MOMMS Family of Matrix Multiplication Algorithms
As the ratio between the rate of computation and rate with which data ca...
research
∙
01/17/2019
Supporting mixed-datatype matrix multiplication within the BLIS framework
We approach the problem of implementing mixed-datatype support within th...
research
∙
08/24/2018
Implementing Strassen's Algorithm with CUTLASS on NVIDIA Volta GPUs
Conventional GPU implementations of Strassen's algorithm (Strassen) typi...
research
∙
08/20/2018
A Simple Methodology for Computing Families of Algorithms
Discovering "good" algorithms for an operation is often considered an ar...
research
∙
10/11/2017
Deriving Correct High-Performance Algorithms
Dijkstra observed that verifying correctness of a program is difficult a...
research
∙
04/11/2017
Strassen's Algorithm for Tensor Contraction
Tensor contraction (TC) is an important computational kernel widely used...
research
∙
11/03/2016
Generating Families of Practical Fast Matrix Multiplication Algorithms
Matrix multiplication (GEMM) is a core operation to numerous scientific ...
research
∙
09/01/2016
BLISlab: A Sandbox for Optimizing GEMM
Matrix-matrix multiplication is a fundamental operation of great importa...
research
∙
05/03/2016