Understanding Performance of eBPF Maps

被引:0
|
作者
Liu, Chang [1 ]
Tak, Byungchul [2 ]
Wang, Long [1 ,3 ]
机构
[1] Tsinghua Univ, Beijing, Peoples R China
[2] Kyungpook Natl Univ, Daegu, South Korea
[3] Zhongguancun Lab, Beijing, Peoples R China
关键词
eBPF; performance benchmark; eBPF map;
D O I
10.1145/3672197.3673430
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
The Linux community has witnessed the rapid development of eBPF technology that allows users to load custom programs into the Linux kernel to extend its capabilities. A key feature that makes eBPF powerful is eBPF maps, which provide data storage and communication capabilities for eBPF programs. However, despite being widely used in eBPF programs, the performance of eBPF maps has received little attention. To understand the performance characteristics of eBPF maps, we conduct a comprehensive benchmark on them. The benchmark results demonstrate the access overhead of different types of eBPF maps and reveal the impact of various factors on the access overhead. By analyzing the benchmark results, we derive some implications for eBPF users to use eBPF maps more efficiently.
引用
收藏
页码:9 / 15
页数:7
相关论文
共 50 条
  • [1] Understanding the Security of Linux eBPF Subsystem
    Mohamed, Mohamed Husain Noor
    Wang, Xiaoguang
    Ravindran, Binoy
    PROCEEDINGS OF THE 14TH ACM SIGOPS ASIA-PACIFIC WORKSHOP ON SYSTEMS, APSYS 2023, 2023, : 87 - 92
  • [2] Performance Implications of Packet Filtering with Linux eBPF
    Scholz, Dominik
    Raumer, Daniel
    Emmerich, Paul
    Kurtz, Alexander
    Lesiak, Krzysztof
    Carle, Georg
    PROCEEDINGS OF THE 2018 30TH INTERNATIONAL TELETRAFFIC CONGRESS (ITC 30), VOL 1, 2018, : 209 - 217
  • [3] The rise of eBPF for non-intrusive performance monitoring
    Cassagnes, Cyril
    Trestioreanu, Lucian
    Joly, Clement
    State, Radu
    NOMS 2020 - PROCEEDINGS OF THE 2020 IEEE/IFIP NETWORK OPERATIONS AND MANAGEMENT SYMPOSIUM 2020: MANAGEMENT IN THE AGE OF SOFTWARIZATION AND ARTIFICIAL INTELLIGENCE, 2020,
  • [4] AI & eBPF based performance anomaly detection system
    Ben-Yair, Ido
    Rogovoy, Pavel
    Zaidenberg, Nezer
    SYSTOR '19: PROCEEDINGS OF THE 12TH ACM INTERNATIONAL SYSTEMS AND STORAGE CONFERENCE, 2019, : 180 - 180
  • [5] Performance evaluation of portable time synchronization method using eBPF
    Kobori, Kazuto
    Lee, Chunghan
    Hirotsu, Toshio
    CONCURRENCY AND COMPUTATION-PRACTICE & EXPERIENCE, 2024, 36 (08):
  • [6] The Road to Understanding Maps
    Liben, Lynn S.
    CURRENT DIRECTIONS IN PSYCHOLOGICAL SCIENCE, 2009, 18 (06) : 310 - 315
  • [7] UNDERSTANDING MAPS - HODGKISS,A
    KEATES, JS
    SCOTTISH GEOGRAPHICAL MAGAZINE, 1982, 98 (01): : 63 - 63
  • [8] UNDERSTANDING THEMATIC MAPS
    MOSENTHAL, PB
    KIRSCH, IS
    JOURNAL OF READING, 1990, 34 (02): : 136 - 140
  • [9] eBPF-sec: A Defensive Framework Against eBPF Attacks on Containers
    Xu, Kejin
    Wang, Xi
    Li, Lun
    Gao, Jin
    2024 IEEE SYMPOSIUM ON COMPUTERS AND COMMUNICATIONS, ISCC 2024, 2024,
  • [10] eWAPA: An eBPF-based WASI Performance Analysis Framework for WebAssembly Runtimes
    Mao, Chenxi
    Su, Yuxin
    Shan, Shiwen
    Li, Dan
    2024 IEEE INTERNATIONAL CONFERENCE ON SOFTWARE SERVICES ENGINEERING, SSE 2024, 2024, : 323 - 333