7 papers
Data-Driven Dynamic Algorithm Dispatch with Large Language Models
Rushil Shah, Emmanuel Lujan, Rabab Alomairy +1
We introduce a large language model (LLM)-driven approach for generating dynamic algorithmic dispatch heuristics in high-performance linear algebra. By combining prompt engineering…
Portable to Efficient: Auto-Tuning Hardware-Agnostic GPU Kernels in Julia
Floris-Jan Willemsen, Evelyne Ringoot, Alan Edelman
Traditionally, GPU kernels have been developed and optimized within vendor-specific programming models to achieve high performance, resulting in software that is difficult to optim…
Cross-Model Cross-Language AI Coding Agent Performance: Accuracy and Speed of Parallel CLRS Algorithms
Shiqi Cheng, Evelyne Ringoot, Rabab Alomairy +1
AI coding agents have quickly become omnipresent in software engineering. Their serial performance, both in terms of accuracy and speed, has been extensively covered. However, rece…
Accelerating Bidiagonalization of Banded Matrices through Memory-Aware Bulge-Chasing on GPUs
Evelyne Ringoot, Rabab Alomairy, Alan Edelman
The reduction of a banded matrix to bidiagonal form is a critical step in the calculation of Singular Values, a cornerstone of scientific computing and AI. Although inherently para…
Hierarchical Recursive Precision for Accelerating Symmetric Linear Solves on MXUs
Vicki Carrica, Rabab Alomairy, Evelyne Ringoot +1
Symmetric positive-definite system solvers based on Cholesky factorization are fundamental to many scientific applications, such as climate modeling. We present a portable, nested…
Performant Unified GPU Kernels for Portable Singular Value Computation Across Hardware and Precision
Evelyne Ringoot, Rabab Alomairy, Valentin Churavy +1
This paper presents a portable, GPU-accelerated implementation of a QR-based singular value computation algorithm in Julia. The singular value ecomposition (SVD) is a fundamental n…