Squeezing Operator Performance Potential for the Ascend Architecture
Open Access
- 30 March 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1156-1171
Abstract
No abstract availableFunding Information
- National Natural Science Foundation of China (62325205, 62172204)
- Nanjing University - China Mobile Communications Group Co., Ltd. Joint Institute
- Natural Science Foundation of Jiangsu Province (BK20243053)
This publication has 25 references indexed in Scilit:
- AIbench: a tool for benchmarking Huawei ascend AI processorsCCF Transactions on High Performance Computing, 2024
- Evaluating performance of AI operators using roofline modelApplied Intelligence, 2021
- The Design Process for Google's Training Chips: TPUv2 and TPUv3IEEE Micro, 2021
- Ascend: a Scalable and Unified Architecture for Ubiquitous Deep Neural Network Computing : Industry Track Paper2021
- MobileNetV32021
- DaVinci: A Scalable Architecture for Neural Network Computing2019
- Benchmark Analysis of Representative Deep Neural Network ArchitecturesIEEE Access, 2018
- Xception: Deep Learning with Depthwise Separable Convolutions2017
- In-Memory Data Rearrangement for Irregular, Data-Intensive ComputingComputer, 2015
- The transglycosylation reaction of cyclodextrin glucanotransferase is operated by a Ping‐Pong mechanismFEBS Letters, 1994