Direct GPU-FPGA Communication Direkte GPU-FPGA

Heterogeneous computing systems consisting of CPUs, GPUs and FPGAs currently suffer from a comparatively low bandwidth and high latency for data transfers between the GPU and the FPGA. So far, no standard or vendor-provided method exists for direct communication between these two devices. Indirect communication with a round-trip via the CPU is required. This thesis describes an example effort to enable this missing link for use with the popular computing platform OpenCL. As expected, a significant increase in bandwidth has been achieved. However, only the direction from the FPGA to the GPU could be realized. More investigation or a different approach is still required to enable the opposite direction as well.

[1]  Rob Williams,et al.  Linux device drivers , 2006 .

[2]  Scott Hauck,et al.  Reconfigurable Computing: The Theory and Practice of FPGA-Based Computation , 2007 .

[3]  David Weimer.  Bibliography , 2018, Medical History. Supplement.

[4]  Implementing FPGA Design with the OpenCL Standard , 2010 .

[5]  Ray Bittner.  Speedy bus mastering PCI express , 2012, 22nd International Conference on Field Programmable Logic and Applications (FPL).

[6]  Ra Inta,et al.  The "Chimera": An Off-The-Shelf CPU/GPGPU/FPGA Hybrid Computing Platform , 2012, Int. J. Reconfigurable Comput..

[7]  Alessandro Forin,et al.  Direct GPU/FPGA communication Via PCI express , 2012, 2012 41st International Conference on Parallel Processing Workshops.

[8]  Zied Marrakchi,et al.  Tree-based Heterogeneous FPGA Architectures: Application Specific Exploration and Optimization , 2012 .

[9]  Shinpei Kato,et al.  Gdev: First-Class GPU Resource Management in the Operating System , 2012, USENIX Annual Technical Conference.

[10]  Nicholas Wilt,et al.  The CUDA Handbook: A Comprehensive Guide to GPU Programming , 2013 .

[11]  Yann Thoma,et al.  FPGA2: An open source framework for FPGA-GPU PCIe communication , 2013, 2013 International Conference on Reconfigurable Computing and FPGAs (ReConFig).

[12]  Nikhil Madduri.  Hardware Accelerated Particle Filter for Lane Detection and Tracking in OpenCL , 2014 .

[13]  Phillipp Bergmann,et al.  Pci Express System Architecture , 2016 .