The PEPPHER composition tool: performance-aware composition for GPU-based systems

被引:8
|
作者
Dastgeer, Usman [1 ]
Li, Lu [1 ]
Kessler, Christoph [1 ]
机构
[1] Linkoping Univ, Dept Comp & Informat Sci, PELAB, S-58183 Linkoping, Sweden
关键词
PEPPHER project; Component model; GPU-based systems; Performance portability; Dynamic scheduling;
D O I
10.1007/s00607-013-0371-8
中图分类号
TP301 [理论、方法];
学科分类号
081202 ;
摘要
The PEPPHER (EU FP7 project) component model defines the notion of component, interface and meta-data for homogeneous and heterogeneous parallel systems. In this paper, we describe and evaluate the PEPPHER composition tool, which explores the application's components and their implementation variants, generates the necessary low-level code that interacts with the runtime system, and coordinates the native compilation and linking of the various code units to compose the overall application code to optimize performance. We discuss the concept of smart containers and its benefits for reducing dispatch overhead, exploiting implicit parallelism across component invocations and runtime optimization of data transfers. In an experimental evaluation with several applications, we demonstrate that the composition tool provides a high-level programming front-end while effectively utilizing the task-based PEPPHER runtime system (StarPU) underneath for different usage scenarios on GPU-based systems.
引用
收藏
页码:1195 / 1211
页数:17
相关论文
共 50 条
  • [41] Smart Containers and Skeleton Programming for GPU-Based Systems
    Usman Dastgeer
    Christoph Kessler
    International Journal of Parallel Programming, 2016, 44 : 506 - 530
  • [42] Smart Containers and Skeleton Programming for GPU-Based Systems
    Dastgeer, Usman
    Kessler, Christoph
    INTERNATIONAL JOURNAL OF PARALLEL PROGRAMMING, 2016, 44 (03) : 506 - 530
  • [43] Performance Prediction of GPU-based Deep Learning Applications
    Gianniti, Eugenio
    Zhang, Li
    Ardagna, Danilo
    CLOSER: PROCEEDINGS OF THE 9TH INTERNATIONAL CONFERENCE ON CLOUD COMPUTING AND SERVICES SCIENCE, 2019, : 279 - 286
  • [44] ProFuN TG: A Tool for Programming and Managing Performance-Aware Sensor Network Applications
    Elsts, Atis
    Bijarbooneh, Farshid Hassani
    Jacobsson, Martin
    Sagonas, Konstantinos
    2015 IEEE 40TH LOCAL COMPUTER NETWORKS CONFERENCE WORKSHOPS (LCN WORKSHOPS), 2015, : 751 - 759
  • [45] Performance-aware deployment of streaming applications in distributed stream computing systems
    Sun, Dawei
    Gao, Shang
    Liu, Xunyun
    Li, Fengyun
    Buyya, Rajkumar
    INTERNATIONAL JOURNAL OF BIO-INSPIRED COMPUTATION, 2020, 15 (01) : 52 - 62
  • [46] PAS: Performance-Aware Job Scheduling for Big Data Processing Systems
    Li, Yiren
    Li, Tieke
    Shen, Pei
    Hao, Liang
    Yang, Jin
    Zhang, Zhengtong
    Chen, Junhao
    Bao, Liang
    SECURITY AND COMMUNICATION NETWORKS, 2022, 2022
  • [47] Cache-Aware Kernel Tiling: An Approach for System-Level Performance Optimization of GPU-Based Applications
    Maghazeh, Arian
    Chattopadhyay, Sudipta
    Eles, Petru
    Peng, Zebo
    2019 DESIGN, AUTOMATION & TEST IN EUROPE CONFERENCE & EXHIBITION (DATE), 2019, : 570 - 575
  • [48] A GPU-Based Monte Carlo Tool for Computing DRRs with Multiple Scattering
    Jia, X.
    Folkerts, M.
    Choi, D.
    Majumdar, A.
    Jiang, S.
    MEDICAL PHYSICS, 2011, 38 (06)
  • [49] GPU-Based Timing-Aware Test Generation for Small Delay Defects
    Liao, Kuan-Yu
    Chen, Po-Juei
    Lin, Ang-Feng
    Li, James Chien-Mo
    Hsiao, Michael S.
    Wang, Laung-Terng
    2014 19TH IEEE EUROPEAN TEST SYMPOSIUM (ETS 2014), 2014,
  • [50] Energy- and Performance-Aware Scheduling of Tasks on Parallel and Distributed Systems
    Sheikh, Hafiz Fahad
    Tan, Hengxing
    Ahmad, Ishfaq
    Ranka, Sanjay
    Bv, Phanisekhar
    ACM JOURNAL ON EMERGING TECHNOLOGIES IN COMPUTING SYSTEMS, 2012, 8 (04)