Direct GPU/FPGA communication Via PCI express

We describe a mechanism for connecting GPU and FPGA devices directly via the PCI Express bus, enabling the transfer of data between these heterogeneous computing units without the intermediate use of system memory. We evaluate the performance benefits of this approach over a range of transfer sizes,...

Full description

Saved in:
Bibliographic Details
Published inCluster computing Vol. 17; no. 2; pp. 339 - 348
Main Authors Bittner, Ray, Ruf, Erik, Forin, Alessandro
Format Journal Article
LanguageEnglish
Published Boston Springer US 01.06.2014
Springer Nature B.V
Subjects
Bandwidths
Bus interconnections
Communication
Computer Communication Networks
Computer Science
Computer vision
Data transfer (computers)
Design
Field programmable gate arrays
Operating Systems
Performance evaluation
Processor Architectures
CUDA
Computer vision
PCI express
FPGA
Verilog
nVidia
PCIe
Windows
Xilinx
GPU
Hand tracking
Online AccessGet full text

Cover

More Information
Summary:We describe a mechanism for connecting GPU and FPGA devices directly via the PCI Express bus, enabling the transfer of data between these heterogeneous computing units without the intermediate use of system memory. We evaluate the performance benefits of this approach over a range of transfer sizes, and demonstrate its utility in a computer vision application. We find that bypassing system memory yields improvements as high as 2.2× in data transfer speed, and 1.9× in application performance.
ISSN:1386-7857
1573-7543
DOI:10.1007/s10586-013-0280-9