RosebudVirt: A High-Performance and Partially Reconfigurable FPGA Virtualization Framework for Multitenant Networks

被引:1
|
作者
Chang, Yiwei [1 ,2 ]
Guo, Zhichuan [1 ,2 ]
机构
[1] Chinese Acad Sci, Inst Acoust, Natl Network New Media Engn Res Ctr, Beijing 100190, Peoples R China
[2] Univ Chinese Acad Sci, Sch Elect Elect & Commun Engn, Beijing 100049, Peoples R China
关键词
Field programmable gate arrays; Virtualization; Hardware; Resource management; Cloud computing; Throughput; Software; Cloud data centers; field-programmable gate array (FPGA) virtualization; multitenant networks; partial reconfiguration (PR); single-root I/O virtualization (SR-IOV);
D O I
10.1109/TVLSI.2024.3436017
中图分类号
TP3 [计算技术、计算机技术];
学科分类号
0812 ;
摘要
Field-programmable gate arrays (FPGAs) are key accelerators in cloud data centers due to their parallelism and programmability. However, challenges such as low hardware utilization and high virtualization overhead persist. This brief presents RosebudVirt, a high-performance and partially reconfigurable FPGA virtualization framework tailored for multitenant networks. It enhances the original Rosebud by introducing single-root I/O virtualization (SR-IOV) support, partitioning the PCIe-attached FPGA device into multiple physical functions (PFs) and virtual functions (VFs) accessible to the linux kernel via PF and VF drivers. This facilitates direct mapping among tenants, VFs, and reconfigurable packet-processing units (RPUs) within the FPGA. RosebudVirt achieves near-native throughput with < 1% area overhead and increases hardware utilization by up to 7.6 times by additional VF drivers and network interfaces. What is more, RosebudVirt is compatible with Kubernetes and Docker
引用
收藏
页码:298 / 302
页数:5
相关论文
共 50 条
  • [21] A Reconfigurable Framework for Performance Enhancement With Dynamic FPGA Configuration Prefetching
    Lifa, Adrian
    Eles, Petru
    Peng, Zebo
    IEEE TRANSACTIONS ON COMPUTER-AIDED DESIGN OF INTEGRATED CIRCUITS AND SYSTEMS, 2016, 35 (01) : 100 - 113
  • [22] Power-efficient and High-performance Block I/O Framework for Mobile Virtualization Systems
    Lee, Kihong
    Lee, Dongwoo
    Eom, Young Ik
    ACM IMCOM 2015, PROCEEDINGS, 2015,
  • [23] Power-efficient and high-performance block I/O framework for mobile virtualization systems
    Kihong Lee
    DongWoo Lee
    Sungkil Lee
    Young Ik Eom
    The Journal of Supercomputing, 2017, 73 : 1307 - 1321
  • [24] Power-efficient and high-performance block I/O framework for mobile virtualization systems
    Lee, Kihong
    Lee, DongWoo
    Lee, Sungkil
    Eom, Young Ik
    JOURNAL OF SUPERCOMPUTING, 2017, 73 (04): : 1307 - 1321
  • [25] A high-performance fully reconfigurable FPGA-based 2D convolution processor
    Perri, S
    Lanuzza, M
    Corsonello, P
    Cocorullo, G
    MICROPROCESSORS AND MICROSYSTEMS, 2005, 29 (8-9) : 381 - 391
  • [26] High-Performance Reconfigurable Computer Systems
    Dordopulo, Alexey
    Kalyaev, Igor
    Levin, Ilya
    Slasten, Liubov
    PARALLEL COMPUTING TECHNOLOGIES, 2011, 6873 : 272 - 283
  • [27] The promise of high-performance reconfigurable computing
    El-Ghazawi, Tarek
    El-Araby, Esam
    Huang, Miaoqing
    Gaj, Kris
    Kindratenko, Volodymyr
    Buell, Duncan
    COMPUTER, 2008, 41 (02) : 69 - +
  • [28] Elements of High-Performance Reconfigurable Computing
    Vancourt, Tom
    Herbordt, Martin C.
    ADVANCES IN COMPUTERS, VOL 75: COMPUTER PERFORMANCE ISSUES, 2009, 75 : 113 - 157
  • [29] A Unified FPGA Virtualization Framework for General-Purpose Deep Neural Networks in the Cloud
    Zeng, Shulin
    Dai, Guohao
    Sun, Hanbo
    Liu, Jun
    Li, Shiyao
    Ge, Guangjun
    Zhong, Kai
    Guo, Kaiyuan
    Wang, Yu
    Yang, Huazhong
    ACM TRANSACTIONS ON RECONFIGURABLE TECHNOLOGY AND SYSTEMS, 2022, 15 (03)
  • [30] Toward High-Performance and Scalable Network Functions Virtualization
    Wang, Chengwei
    Spatscheck, Oliver
    Gopalakrishnan, Vijay
    Xu, Yang
    Applegate, David
    IEEE INTERNET COMPUTING, 2016, 20 (06) : 10 - 20