5 papers
AI Benchmark Democratization and Carpentry
Gregor von Laszewski, Wesley Brewer, Jeyan Thiyagalingam +28
Benchmarks are a cornerstone of modern machine learning, enabling reproducibility, comparison, and scientific progress. However, AI benchmarks are increasingly complex, requiring d…
HPL-MxP Benchmark: Mixed-Precision Algorithms, Iterative Refinement, and Scalable Data Generation
Jack Dongarra, Piotr Luszczek
We present a mixed-precision benchmark called HPL-MxP that uses both a lower-precision LU factorization with a non-stationary iterative refinement based on GMRES. We evaluate the n…
The Stability of Block Eliminations and Additive Modifications
Neil Lindquist, Piotr Luszczek, Jack Dongarra
The block elimination with additive modifications (BEAM) method was recently proposed as a alternative to LU with partial pivoting requiring less communication. Because of the nove…
CholeskyQR with Randomization and Pivoting for Tall Matrices (CQRRPT)
Maksim Melnichenko, Oleg Balabanov, Riley Murray +3
This paper develops and analyzes a new algorithm for QR decomposition with column pivoting (QRCP) of rectangular matrices with many more rows than columns. The algorithm carefully…
Interface for Sparse Linear Algebra Operations
Ahmad Abdelfattah, Willow Ahrens, Hartwig Anzt +32
The standardization of an interface for dense linear algebra operations in the BLAS standard has enabled interoperability between different linear algebra libraries, thereby boosti…