Multi-Parameter Performance Modeling Based on Machine Learning with Basic Block Features

被引:0
|
作者
Hao, Meng [1 ]
Zhang, Weizhe [1 ,2 ]
Wang, Yiming [1 ]
Li, Dong [3 ]
Xia, Wen [2 ]
Wang, Hao [4 ]
Lou, Chen [1 ]
机构
[1] Harbin Inst Technol, Sch Comp Sci & Technol, Harbin, Peoples R China
[2] Peng Cheng Lab, Cyberspace Secur Res Ctr, Shenzhen, Peoples R China
[3] Univ Calif, Dept Elect Engn & Comp Sci, Merced, CA USA
[4] Norwegian Univ Sci & Technol, Dept Comp Sci, Trondheim, Norway
来源
2019 IEEE INTL CONF ON PARALLEL & DISTRIBUTED PROCESSING WITH APPLICATIONS, BIG DATA & CLOUD COMPUTING, SUSTAINABLE COMPUTING & COMMUNICATIONS, SOCIAL COMPUTING & NETWORKING (ISPA/BDCLOUD/SOCIALCOM/SUSTAINCOM 2019) | 2019年
基金
中国国家自然科学基金;
关键词
performance modeling; parallel application; basic block feature; machine learning;
D O I
10.1109/ISPA-BDCloud-SustainCom-SocialCom48970.2019.00054
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Considering the increasing complexity and scale of HPC architecture and software, the performance modeling of parallel applications on large-scale HPC platforms has become increasingly important. It plays an important role in many areas, such as performance analysis, job management, and resource estimation. In this work, we propose a multi-parameter performance modeling and prediction framework called MPerfPred, which utilizes basic block frequencies as features and uses machine learning algorithms to automatically construct multi-parameter performance models with high generalization ability. To reduce the prediction overhead, we propose some feature-filtering strategies to reduce the number of features in the training stage and build a serial program called BBF collector for each target application to quickly collect feature values in the prediction stage. We demonstrate the use of MPerfPred on the TianHe-2 supercomputer with six parallel applications. Results show that MPerfPred with SVR achieves better prediction than other input parameter-based modeling methods. The average prediction error and average standard deviation of prediction errors of MPerfPred are 8.42% and 6.09%, respectively. In the prediction stage, the average prediction overhead of MPerfPred is less than 0.13% of the total execution time.
引用
收藏
页码:316 / 323
页数:8
相关论文
共 50 条
  • [1] Optical Fiber Multi-Parameter Measurement Based on Machine Learning
    Ma Zehang
    Gong Rui
    Li Bin
    Pei Li
    Wei Huai
    ACTA OPTICA SINICA, 2022, 42 (20)
  • [2] Fast Multi-Parameter Performance Modeling
    Calotoiu, Alexandru
    Beckingsale, David
    Earl, Christopher W.
    Hoefler, Torsten
    Karlin, Ian
    Schulz, Martin
    Wolf, Felix
    2016 IEEE INTERNATIONAL CONFERENCE ON CLUSTER COMPUTING (CLUSTER), 2016, : 172 - 181
  • [3] Machine Learning Based Multi-Parameter Modeling for Prediction of Post-Inflammatory Lung Changes
    Widmann, Gerlig
    Luger, Anna Katharina
    Sonnweber, Thomas
    Schwabl, Christoph
    Cima, Katharina
    Gerstner, Anna Katharina
    Pizzini, Alex
    Sahanic, Sabina
    Boehm, Anna
    Coen, Maxmilian
    Woell, Ewald
    Weiss, Guenter
    Kirchmair, Rudolf
    Gruber, Leonhard
    Feuchtner, Gudrun M.
    Tancevski, Ivan
    Loeffler-Ragg, Judith
    Tymoszuk, Piotr
    DIAGNOSTICS, 2025, 15 (06)
  • [4] Surface roughness measurement method based on multi-parameter modeling learning
    Chen, Suting
    Feng, Rui
    Zhang, Chuang
    Zhang, Yanyan
    MEASUREMENT, 2018, 129 : 664 - 676
  • [5] An Interpolation-Based Approach to Multi-Parameter Performance Modeling for Heterogeneous Systems
    Rudolph, Dylan
    Stitt, Greg
    PROCEEDINGS OF THE ASAP2015 2015 IEEE 26TH INTERNATIONAL CONFERENCE ON APPLICATION-SPECIFIC SYSTEMS, ARCHITECTURES AND PROCESSORS, 2015, : 174 - 180
  • [6] A multi-parameter control method for maize threshing based on machine learning algorithm optimisation
    Fan, Chenlong
    Zhang, Dongxing
    Yang, Li
    Cui, Tao
    He, Xiantao
    Qiao, Mengmeng
    Sun, Jialu
    Dong, Jiaqi
    BIOSYSTEMS ENGINEERING, 2023, 236 : 212 - 223
  • [7] The multi-parameter combination mind-evolutionary-based machine learning and its application
    Xie, KM
    Mou, CH
    Xie, G
    SMC 2000 CONFERENCE PROCEEDINGS: 2000 IEEE INTERNATIONAL CONFERENCE ON SYSTEMS, MAN & CYBERNETICS, VOL 1-5, 2000, : 183 - 187
  • [8] Machine Learning-Based Precursor Detection Using Seismic Multi-Parameter Data
    Lu, Xian
    Wang, Qiong
    Zhang, Xiaodong
    Yan, Wei
    Meng, Lingyuan
    Wang, Haitao
    APPLIED SCIENCES-BASEL, 2024, 14 (06):
  • [9] Multi-parameter machine learning approach to the neuroanatomical basis of developmental dyslexia
    Plonski, Piotr
    Gradkowski, Wojciech
    Altarelli, Irene
    Monzalvo, Karla
    van Ermingen-Marbach, Muna
    Grande, Marion
    Heim, Stefan
    Marchewka, Artur
    Bogorodzki, Piotr
    Ramus, Franck
    Jednorog, Katarzyna
    HUMAN BRAIN MAPPING, 2017, 38 (02) : 900 - 908
  • [10] Application of Machine Learning Models to Multi-Parameter Maximum Magnitude Prediction
    Zhang, Jingye
    Sun, Ke
    Han, Xiaoming
    Mao, Ning
    APPLIED SCIENCES-BASEL, 2024, 14 (24):