Performance Prediction of GPU-based Deep Learning Applications

被引：5
|
作者：
Gianniti, Eugenio [1 ]
Zhang, Li [2 ]
Ardagna, Danilo [1 ]
机构：
[1] Politecn Milan, Dip Elettron Informaz & Bioingn, Milan, Italy
[2] IBM Res, Yorktown Hts, NY USA
来源：
CLOSER: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING AND SERVICES SCIENCE | 2019年
基金：
欧盟地平线“2020”;
关键词：
Convolutional Neural Networks; Deep Learning; Performance Prediction; General Purpose GPUs; MODEL;
D O I：
10.5220/0007681802790286
中图分类号：
TP301 [理论、方法];
学科分类号：
081202 ;
摘要：
Recent years saw an increasing success in the application of deep learning methods across various domains and for tackling different problems, ranging from image recognition and classification to text processing and speech recognition. In this paper we propose, discuss, and validate a black box approach to model the execution time for training convolutional neural networks (CNNs), with a particular focus on deployments on general purpose graphics processing units (GPGPUs). We demonstrate that our approach is generally applicable to a variety of CNN models and different types of GPGPUs with high accuracy. The proposed method can support with great precision (within 5% average percentage error) the management of production environments.
引用
收藏
页码：279 / 286
页数：8
相关论文
共 50 条
  • [1] Performance Prediction of GPU-based Deep Learning Applications
    Gianniti, Eugenio
    Zhang, Li
    Ardagna, Danilo
    2018 30TH INTERNATIONAL SYMPOSIUM ON COMPUTER ARCHITECTURE AND HIGH PERFORMANCE COMPUTING (SBAC-PAD 2018), 2018, : 167 - 170
  • [2] DeepMon: Mobile GPU-based Deep Learning Framework for Continuous Vision Applications
    Huynh, Loc N.
    Lee, Youngki
    Balan, Rajesh Krishna
    MOBISYS'17: PROCEEDINGS OF THE 15TH ANNUAL INTERNATIONAL CONFERENCE ON MOBILE SYSTEMS, APPLICATIONS, AND SERVICES, 2017, : 82 - 95
  • [3] Scheduling CPU for GPU-based Deep Learning Jobs
    Xiao, Wencong
    Han, Zhenhua
    Zhao, Hanyu
    Peng, Xuan
    Zhang, Quanlu
    Yang, Fan
    Zhou, Lidong
    PROCEEDINGS OF THE 2018 ACM SYMPOSIUM ON CLOUD COMPUTING (SOCC '18), 2018, : 503 - 503
  • [4] Performance prediction of deep learning applications training in GPU as a service systems
    Lattuada, Marco
    Gianniti, Eugenio
    Ardagna, Danilo
    Zhang, Li
    CLUSTER COMPUTING-THE JOURNAL OF NETWORKS SOFTWARE TOOLS AND APPLICATIONS, 2022, 25 (02): : 1279 - 1302
  • [5] Performance prediction of deep learning applications training in GPU as a service systems
    Marco Lattuada
    Eugenio Gianniti
    Danilo Ardagna
    Li Zhang
    Cluster Computing, 2022, 25 : 1279 - 1302
  • [6] Advancing drug discovery via GPU-based deep learning
    Gawehn, Erik
    Hiss, Jan A.
    Brown, J. B.
    Schneider, Gisbert
    EXPERT OPINION ON DRUG DISCOVERY, 2018, 13 (07) : 579 - 582
  • [7] An Efficient and Accurate GPU-based Deep Learning Model for Multimedia Recommendation
    Djenouri, Youcef
    Belhadi, Asma
    Srivastava, Gautam
    Lin, Jerry Chun-Wei
    ACM TRANSACTIONS ON MULTIMEDIA COMPUTING COMMUNICATIONS AND APPLICATIONS, 2024, 20 (02)
  • [8] Comparison of Deep Learning in Neural Networks on CPU and GPU-based frameworks
    Aida-zade, Kamil
    Mustafayev, Elshan
    Rustamov, Samir
    2017 11TH IEEE INTERNATIONAL CONFERENCE ON APPLICATION OF INFORMATION AND COMMUNICATION TECHNOLOGIES (AICT 2017), 2017, : 95 - 98
  • [9] High performance GPU-based parity computing scheduler in storage applications
    Pirahandeh, Mehdi
    Kim, Deok-Hwan
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2017, 29 (04):
  • [10] A Framework for Performance-aware Composition of Applications for GPU-based Systems
    Dastgeer, Usman
    Kessler, Christoph
    2013 42ND ANNUAL INTERNATIONAL CONFERENCE ON PARALLEL PROCESSING (ICPP), 2013, : 698 - 707
12345