NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing-Reference-Cited by-同舟云学术

NeuPIMs: NPU-PIM Heterogeneous Acceleration for Batched LLM Inferencing

Published:2024-04-27 Issue: Volume:35 Page:722-737
ISSN:
Container-title:Proceedings of the 29th ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 3
language:
Short-container-title:

Author:

Heo Guseul¹^ORCID,Lee Sangyeop²^ORCID,Cho Jaehong¹^ORCID,Choi Hyunmin¹^ORCID,Lee Sanghyeon²^ORCID,Ham Hyungkyu³^ORCID,Kim Gwangsun⁴^ORCID,Mahajan Divya⁵^ORCID,Park Jongse²^ORCID

Affiliation:

1. KAIST, Daejeon, Republic of Korea

2. KAIST, Daejeon, Korea, South ? Republic of Korea

3. POSTECH, Pohang, Republic of Korea

4. POSTECH, Pohang, Korea, South ? Republic of Korea

5. Georgia Institute of Technology, Atlanta, GA, United States of America

Publisher

ACM

Reference89 articles.

1. Nvidia Tensor RT 4.0. https://developer.nvidia.com/tensorrt.

2. HuggingFace. https://github.com/huggingface/transformers/tree/main, 2022.

3. ONNXim NPU Simulator. https://github.com/PSAL-POSTECH/ONNXim, 2024.

4. Accelerating recommendation system training by leveraging popular choices

5. Muhammad Adnan, Yassaman Ebrahimzadeh Maboud, Divya Mahajan, and Prashant J Nair. Heterogeneous Acceleration Pipeline for Recommendation System Training. arXiv preprint arXiv:2204.05436, 2022.

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. pSyncPIM: Partially Synchronous Execution of Sparse Matrix Operations for All-Bank PIM Architectures;2024 ACM/IEEE 51st Annual International Symposium on Computer Architecture (ISCA);2024-06-29