Mosaic: Exploiting Instruction-Level Parallelism on Deep Learning Accelerators with iTex Tessellation
Open Access
- 30 March 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 672-688
Abstract
No abstract availableFunding Information
- The National Key R&D Program of China (2022YFB4501600)
- the Strategic Priority Research Program of the Chinese Academy of Sciences (XDB0660300, XDB0660301, XDB0660302)
- CAS Project for Young Scientists in Basic Research (YSBR-029)
- the NSF of China (U22A2028, 62302483, 62222214, 62341411, 62102399, 62302478, 62302482, 62302480, 62302481)
This publication has 19 references indexed in Scilit:
- Exploiting Intra-SM Parallelism in GPUs via Persistent and Elastic Blocks2021
- MLIR: Scaling Compiler Infrastructure for Domain Specific Computation2021
- Tiramisu: A Polyhedral Compiler for Expressing Fast and Portable Code2019
- NVIDIA Tensor Core Programmability, Performance & Precision2018
- Efficient Processing of Deep Neural Networks: A Tutorial and SurveyProceedings of the IEEE, 2017
- In-Datacenter Performance Analysis of a Tensor Processing Unit2017
- GPU register file virtualization2015
- Halide2013
- Neural Acceleration for General-Purpose Approximate Programs2012
- Reducibility among Combinatorial Problems1972