A unified, low-overhead framework to support continuous profiling and optimization

被引:0
|
作者
Zhang, M [1 ]
He, XB [1 ]
Yang, Q [1 ]
机构
[1] Univ Rhode Isl, Kingston, RI 02881 USA
关键词
embedded system; continuous profiling; on-line optimization; performance evaluation;
D O I
10.1109/PCCC.2003.1203715
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
We propose a unified, low-overhead framework (ULF) to support continuous system profiling and optimization based on a specifically designed embedded board. Instead of building a new profiling tool from scratch, ULF provides a unified interface to integrate various existing profiling tools and optimizers, and helps to easily build future tools. ULF uses an embedded processor to off load tasks of post-processing profiling data, which reduces system overhead caused by profiling tools and makes ULF especially suitable for continuous profiling on production systems. By processing the profiling data in parallel and providing feedback promptly, ULF supports on-line optimization. Our case study on I/O profiling demonstrated that ULF- enhanced profiling tool dramatically reduces overhead making continuous profiling on production systems feasible.
引用
收藏
页码:327 / 334
页数:8
相关论文
共 50 条
  • [1] A Low-Overhead Dynamic Optimization Framework for Multicores
    Fletcher, Christopher W.
    Harding, Rachael
    Khan, Omer
    Devadas, Srinivas
    PROCEEDINGS OF THE 21ST INTERNATIONAL CONFERENCE ON PARALLEL ARCHITECTURES AND COMPILATION TECHNIQUES (PACT'12), 2012, : 467 - 468
  • [2] A low-overhead profiling and visualization framework for Hybrid Transactional Memory
    Arcas, Oriol
    Kirchhofer, Philipp
    Soenmez, Nehir
    Schindewolf, Martin
    Unsal, Osman S.
    Karl, Wolfgang
    Cristal, Adrian
    2012 IEEE 20TH ANNUAL INTERNATIONAL SYMPOSIUM ON FIELD-PROGRAMMABLE CUSTOM COMPUTING MACHINES (FCCM), 2012, : 1 - 8
  • [3] A Low-Overhead Method of Embedded Software Profiling
    Liu Fagui
    Li Shengwen
    Xie Ran
    Luo Chunwei
    2009 ISECS INTERNATIONAL COLLOQUIUM ON COMPUTING, COMMUNICATION, CONTROL, AND MANAGEMENT, VOL IV, 2009, : 436 - 439
  • [4] Low-Overhead Reuse Distance Profiling Tool for Multicore
    Sasongko, Muhammad Aditya
    Chabbi, Milind
    Unat, Didem
    EURO-PAR 2021: PARALLEL PROCESSING WORKSHOPS, 2022, 13098 : 555 - 559
  • [5] Low-overhead decision support for dynamic buffer reallocation
    Schmidt, Karsten
    Baechle, Sebastian
    COMPUTER SCIENCE-RESEARCH AND DEVELOPMENT, 2012, 27 (01): : 29 - 43
  • [6] A Low-Overhead Framework for Inexpensive Embedded Control Systems
    Bertolotti, Ivan Cibrario
    Hu, Tingting
    Kashani, Gilda Ghafour Zadeh
    TWELFTH INTERNATIONAL CONFERENCE ON DIGITAL TELECOMMUNICATIONS (ICDT 2017), 2017, : 7 - 12
  • [7] Low-Overhead Trace Collection and Profiling on GPU Compute Kernels
    Darche, Sebastien
    Dagenais, Michel R.
    ACM TRANSACTIONS ON PARALLEL COMPUTING, 2024, 11 (02)
  • [8] A low-overhead asynchronous consensus framework for distributed bundle adjustment
    Liu, Zhuo-hao
    Diao, Chang-yu
    Xing, Wei
    Lu, Dong-ming
    FRONTIERS OF INFORMATION TECHNOLOGY & ELECTRONIC ENGINEERING, 2020, 21 (10) : 1442 - 1454
  • [9] Low-overhead memory leak detection using adaptive statistical profiling
    Chilimbi, TM
    Hauswirth, M
    ACM SIGPLAN NOTICES, 2004, 39 (11) : 156 - 164
  • [10] A Novel Low-Overhead Flexible Instrumentation Framework for Virtual Platforms
    Carmel-Veilleux, Tennessee
    Boland, Jean-Francois
    Bois, Guy
    2011 22ND IEEE INTERNATIONAL SYMPOSIUM ON RAPID SYSTEM PROTOTYPING (RSP), 2011, : 92 - 98