4 papers · 1 filter
tritonBLAS: Triton-based Analytical Approach for GEMM Kernel Parameter Selection
Ryan Swann, Muhammad Osama, Xiaohu Guo +8
We present tritonBLAS, a fast and deterministic analytical model that uses architectural parameters like the cache hierarchy, and relative code and data placement to generate perfo…
Stream-K++: Adaptive GPU GEMM Kernel Scheduling and Selection using Bloom Filters
Harisankar Sadasivan, Muhammed Emin Ozturk, Muhammad Osama +7
General matrix multiplication (GEMM) operations are the fundamental building blocks of computational domains including artificial intelligence (AI). As GPU architectures evolve and…
Iris: First-Class Multi-GPU Programming Experience in Triton
Muhammad Awad, Muhammad Osama, Brandon Potter
Multi-GPU programming traditionally requires developers to navigate complex trade-offs between performance and programmability. High-performance implementations typically rely on l…
Eliminating Multi-GPU Performance Taxes: A Systems Approach to Efficient Distributed LLMs
Octavian Alexandru Trifan, Karthik Sangaiah, Muhammad Awad +5
As large language models (LLMs) continue to scale, their workloads increasingly rely on distributed execution across multiple GPUs. However, the conventional bulk synchronous paral…