HeTraX: Energy Efficient 3D Heterogeneous Manycore Architecture for Transformer Acceleration

被引:0
|
作者
Dhingra, Pratyush [1 ]
Doppa, Janardhan Rao [1 ]
Pande, Partha Pratim [1 ]
机构
[1] Washington State Univ, Pullman, WA 99164 USA
来源
PROCEEDINGS OF THE 29TH ACM/IEEE INTERNATIONAL SYMPOSIUM ON LOW POWER ELECTRONICS AND DESIGN, ISLPED 2024 | 2024年
基金
美国国家科学基金会;
关键词
Transformer; Heterogeneity; Accelerator; Thermal-aware; PIM;
D O I
10.1145/3665314.3670814
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Transformers have revolutionized deep learning and generative modeling to enable unprecedented advancements in natural language processing tasks and beyond. However, designing hardware accelerators for executing transformer models is challenging due to the wide variety of computing kernels involved in the transformer architecture. Existing accelerators are either inadequate to accelerate end-to-end transformer models or suffer notable thermal limitations. In this paper, we propose the design of a three-dimensional heterogeneous architecture referred to as HeTraX specifically optimized to accelerate end-to-end transformer models. HeTraX employs hardware resources aligned with the computational kernels of transformers and optimizes both performance and energy. Experimental results show that HeTraX outperforms existing state-of-the-art by up to 5.6x in speedup and improves EDP by 14.5x while ensuring thermally feasibility.
引用
收藏
页数:6
相关论文
共 50 条
  • [1] H3D-Transformer: A Heterogeneous 3D (H3D) Computing Platform for Transformer Model Acceleration on Edge Devices
    Luo, Yandong
    Yu, Shimeng
    ACM TRANSACTIONS ON DESIGN AUTOMATION OF ELECTRONIC SYSTEMS, 2024, 29 (03)
  • [2] HuNT: Exploiting Heterogeneous PIM Devices to Design a 3-D Manycore Architecture for DNN Training
    Ogbogu, Chukwufumnanya
    Narang, Gaurav
    Joardar, Biresh Kumar
    Doppa, Janardhan Rao
    Chakrabarty, Krishnendu
    Pande, Partha Pratim
    IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2024, 43 (11) : 3300 - 3311
  • [3] Transformer: Run-time reprogrammable heterogeneous architecture for transparent acceleration of dynamic workloads
    Li, Peilong
    Luo, Yan
    Yang, Jun
    JOURNAL OF PARALLEL AND DISTRIBUTED COMPUTING, 2015, 86 : 45 - 61
  • [4] FNE-PCT: An Efficient Transformer Network for 3D Classification
    Han, Ming
    Sha, Jianjun
    Wang, Yanheng
    Ma, Chengyuan
    Zhang, Xiang
    PROCEEDINGS OF 2022 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION (IEEE ICMA 2022), 2022, : 1046 - 1050
  • [5] DAST: Differentiable Architecture Search with Transformer for 3D Medical Image Segmentation
    Yang, Dong
    Xu, Ziyue
    He, Yufan
    Nath, Vishwesh
    Li, Wenqi
    Myronenko, Andriy
    Hatamizadeh, Ali
    Zhao, Can
    Roth, Holger R.
    Xu, Daguang
    MEDICAL IMAGE COMPUTING AND COMPUTER ASSISTED INTERVENTION, MICCAI 2023, PT III, 2023, 14222 : 747 - 756
  • [6] Energy-Efficient Architecture for CNNs Inference on Heterogeneous FPGA
    Spagnolo, Fanny
    Perri, Stefania
    Frustaci, Fabio
    Corsonello, Pasquale
    JOURNAL OF LOW POWER ELECTRONICS AND APPLICATIONS, 2020, 10 (01)
  • [7] ETUNet:Exploring efficient transformer enhanced UNet for 3D brain tumor segmentation
    Zhang, Wang
    Chen, Shanxiong
    Ma, Yuqi
    Liu, Yu
    Cao, Xu
    COMPUTERS IN BIOLOGY AND MEDICINE, 2024, 171
  • [8] 3D-ReG: A 3D ReRAM-based Heterogeneous Architecture for Training Deep Neural Networks
    Li, Bing
    Doppa, Janardhan Rao
    Pande, Partha Pratim
    Chakrabarty, Krishnendu
    Qiu, Joe X.
    Li, Hai
    ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS, 2020, 16 (02)
  • [9] Efficient Hierarchical Multi-view Fusion Transformer for 3D Human Pose Estimation
    Zhou, Kangkang
    Zhang, Lijun
    Lu, Feng
    Zhou, Xiang-Dong
    Shi, Yu
    PROCEEDINGS OF THE 31ST ACM INTERNATIONAL CONFERENCE ON MULTIMEDIA, MM 2023, 2023, : 7512 - 7520
  • [10] Efficient combined algorithm of Transformer and U-Net for 3D medical image segmentation
    Zhang, Mingyan
    Wang, Aixia
    Yang, Gang
    Li, Jingjiao
    2023 35TH CHINESE CONTROL AND DECISION CONFERENCE, CCDC, 2023, : 4377 - 4382