Mixed-Precision S/DGEMM Using the TF32 and TF64 Frameworks on Low-Precision AI Tensor Cores-Reference-Cited by-同舟云学术

Mixed-Precision S/DGEMM Using the TF32 and TF64 Frameworks on Low-Precision AI Tensor Cores

Published:2023-11-12 Issue: Volume: Page:
ISSN:
Container-title:Proceedings of the SC '23 Workshops of The International Conference on High Performance Computing, Network, Storage, and Analysis
language:
Short-container-title:

Author:

Valero-Lara Pedro¹^ORCID,Jorquera Ian²^ORCID,Lui Frank³^ORCID,Vetter Jeffrey¹^ORCID

Affiliation:

1. Oak Ridge National Laboratory, United States of America

2. Colorado State University, United States of America

3. Oak Ridge National Laboratory, United States

Funder

Exascale Computing Project

Publisher

ACM

Link

https://dl.acm.org/doi/pdf/10.1145/3624062.3624084

Reference40 articles.

1. 2018. The GCC Quad-Precision Math Library. https://gcc.gnu.org/onlinedocs/gcc-8.2.0/libquadmath.pdf [Online ; accessed 30- May - 2023 ]. 2018. The GCC Quad-Precision Math Library. https://gcc.gnu.org/onlinedocs/gcc-8.2.0/libquadmath.pdf [Online; accessed 30-May-2023].

2. 2021. AMD Instinct MI250X Accelerator. https://www.amd.com/en/products/server-accelerators/instinct-mi250x. [Online ; accessed 30- May - 2023 ]. 2021. AMD Instinct MI250X Accelerator. https://www.amd.com/en/products/server-accelerators/instinct-mi250x. [Online; accessed 30-May-2023].

3. 2022. Cerebras. https://www.cerebras.net/. [Online ; accessed 30- May - 2023 ]. 2022. Cerebras. https://www.cerebras.net/. [Online; accessed 30-May-2023].

4. 2022. The Scalable Matrix Extension (SME), for Armv9-A. https://developer.arm.com/documentation/ddi0616/latest. [Online ; accessed 30- May - 2023 ]. 2022. The Scalable Matrix Extension (SME), for Armv9-A. https://developer.arm.com/documentation/ddi0616/latest. [Online; accessed 30-May-2023].

5. A survey of numerical linear algebra methods utilizing mixed-precision arithmetic

Cited by 1 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. MixPert: Optimizing Mixed-Precision Floating-Point Emulation on GPU Integer Tensor Cores;Proceedings of the 25th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems;2024-06-20