Bridging FPGA and GPU technologies for AO real-time control

被引:13
作者
Perret, Denis [1 ]
Laine, Maxime [1 ]
Bernard, Julien [1 ]
Gratadour, Damien [1 ]
Sevin, Arnaud [1 ]
机构
[1] CNRS, Observ Paris, LESIA, Meudon, France
来源
ADAPTIVE OPTICS SYSTEMS V | 2016年 / 9909卷
关键词
Adaptive Optics; FPGA; GPU; PCIE; 10G Ethernet; RTC; DMA; Peer-to-Peer;
D O I
10.1117/12.2232858
中图分类号
O43 [光学];
学科分类号
070207 ; 0803 ;
摘要
Our team has developed a common environment for high performance simulations and real-time control of AO systems based on the use of Graphics Processors Units in the context of the COMPASS project. Such a solution, based on the ability of the real time core in the simulation to provide adequate computing performance, limits the cost of developing AO RTC systems and makes them more scalable. A code developed and validated in the context of the simulation may be injected directly into the system and tested on sky. Furthermore, the use of relatively low cost components also offers significant advantages for the system hardware platform. However, the use of GPUs in an AO loop comes with drawbacks: the traditional way of offloading computation from CPU to GPUs -involving multiple copies and unacceptable overhead in kernel launching -is not well suited in a real time context. This last application requires the implementation of a solution enabling direct memory access (DMA) to the GPU memory from a third party device, bypassing the operating system. This allows this device to communicate directly with the real-time core of the simulation feeding it with the WFS camera pixel stream. We show that DMA between a custom FPGA-based frame-grabber and a computation unit (GPU, FPGA, or Coprocessor such as Xeon-phi) across PCIe allows us to get latencies compatible with what will be needed on ELTs. As a fine-grained synchronization mechanism is not yet made available by GPU vendors, we propose the use of memory polling to avoid interrupts handling and involvement of a CPU. Network and Vision protocols are handled by the FPGA-based Network Interface Card (NIC). We present the results we obtained on a complete AO loop using camera and deformable mirror simulators.
引用
收藏
页数:11
相关论文
共 50 条
  • [31] A real-time motion estimation FPGA architecture
    Babionitakis, Konstantinos
    Doumenis, Gregory A.
    Georgakarakos, George
    Lentaris, George
    Nakos, Kostantinos
    Reisis, Dionysios
    Sifnaios, Ioannis
    Vlassopoulos, Nikolaos
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2008, 3 (1-2) : 3 - 20
  • [32] FPGA memory optimization for real-time imaging
    Houzet, Dominique
    Fresse, Virginie
    Konik, Hubert
    PROCEEDINGS OF THE 2016 CONFERENCE ON DESIGN AND ARCHITECTURES FOR SIGNAL & IMAGE PROCESSING, 2016, : 176 - 182
  • [33] A real-time motion estimation FPGA architecture
    Konstantinos Babionitakis
    Gregory A. Doumenis
    George Georgakarakos
    George Lentaris
    Kostantinos Nakos
    Dionysios Reisis
    Ioannis Sifnaios
    Nikolaos Vlassopoulos
    Journal of Real-Time Image Processing, 2008, 3 : 3 - 20
  • [34] Real-time Learned Image Codec on FPGA
    Sun, Heming
    Yi, Qingyang
    Lin, Fangzheng
    Yu, Lu
    Katto, Jiro
    Fujita, Masahiro
    2022 IEEE INTERNATIONAL CONFERENCE ON VISUAL COMMUNICATIONS AND IMAGE PROCESSING (VCIP), 2022,
  • [35] Real-Time BIST for Fault Detection in FPGA
    Ju, Xiaoming
    Zhang, Jiehao
    Zhang, Yizhong
    PROCEEDINGS OF THE 2013 INTERNATIONAL CONFERENCE ON INFORMATION SCIENCE AND TECHNOLOGY APPLICATIONS (ICISTA-2013), 2013, 58 : 31 - 36
  • [36] Real-time Shape and Pedestrian Detection with FPGA
    Xiao, Han
    Song, Haitao
    He, Wenhao
    Yuan, Kui
    2015 IEEE INTERNATIONAL CONFERENCE ON MECHATRONICS AND AUTOMATION, 2015, : 2381 - 2386
  • [37] An FPGA implementation for real-time edge detection
    Jiang, Jie
    Liu, Chang
    Ling, Sirui
    JOURNAL OF REAL-TIME IMAGE PROCESSING, 2018, 15 (04) : 787 - 797
  • [38] An FPGA implementation for real-time edge detection
    Jie Jiang
    Chang Liu
    Sirui Ling
    Journal of Real-Time Image Processing, 2018, 15 : 787 - 797
  • [39] Real-time massive convolution for audio applications on GPU Massive convolution on GPU
    Belloch, Jose A.
    Gonzalez, Alberto
    Martinez-Zaldivar, F. J.
    Vidal, Antonio M.
    JOURNAL OF SUPERCOMPUTING, 2011, 58 (03) : 449 - 457
  • [40] A GPU based real-time GPS software receiver
    Thomas Hobiger
    Tadahiro Gotoh
    Jun Amagai
    Yasuhiro Koyama
    Tetsuro Kondo
    GPS Solutions, 2010, 14 : 207 - 216