article yes Tobias Kenter author 3145 Henning Schmitz author Christian Plessl author 161530000-0001-5728-9982 27 department 518 department 78 department SFB 901 project SFB 901 - Subprojekt C2 project SFB 901 - Project Area C project Self-Adaptive Virtualisation-Aware High-Performance/Low-Energy Heterogeneous System Architectures project FPGAs are known to permit huge gains in performance and efficiency for suitable applications but still require reduced design efforts and shorter development cycles for wider adoption. In this work, we compare the resulting performance of two design concepts that in different ways promise such increased productivity. As common starting point, we employ a kernel-centric design approach, where computational hotspots in an application are identified and individually accelerated on FPGA. By means of a complex stereo matching application, we evaluate two fundamentally different design philosophies and approaches for implementing the required kernels on FPGAs. In the first implementation approach, we designed individually specialized data flow kernels in a spatial programming language for a Maxeler FPGA platform; in the alternative design approach, we target a vector coprocessor with large vector lengths, which is implemented as a form of programmable overlay on the application FPGAs of a Convey HC-1. We assess both approaches in terms of overall system performance, raw kernel performance, and performance relative to invested resources. After compensating for the effects of the underlying hardware platforms, the specialized dataflow kernels on the Maxeler platform are around 3x faster than kernels executing on the Convey vector coprocessor. In our concrete scenario, due to trade-offs between reconfiguration overheads and exposed parallelism, the advantage of specialized dataflow kernels is reduced to around 2.5x. https://ris.uni-paderborn.de/download/296/1444/296-859425.pdf application/pdf Hindawi2015 eng 10.1155/2015/859425 2015 T. Kenter, H. Schmitz, C. Plessl, International Journal of Reconfigurable Computing (IJRC) 2015 (2015). Kenter, Tobias, et al. “Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study.” <i>International Journal of Reconfigurable Computing (IJRC)</i>, vol. 2015, 859425, Hindawi, 2015, doi:<a href="https://doi.org/10.1155/2015/859425">10.1155/2015/859425</a>. @article{Kenter_Schmitz_Plessl_2015, title={Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study}, volume={2015}, DOI={<a href="https://doi.org/10.1155/2015/859425">10.1155/2015/859425</a>}, number={859425}, journal={International Journal of Reconfigurable Computing (IJRC)}, publisher={Hindawi}, author={Kenter, Tobias and Schmitz, Henning and Plessl, Christian}, year={2015} } Kenter, T., Schmitz, H., &#38; Plessl, C. (2015). Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study. <i>International Journal of Reconfigurable Computing (IJRC)</i>, <i>2015</i>, Article 859425. <a href="https://doi.org/10.1155/2015/859425">https://doi.org/10.1155/2015/859425</a> Kenter T, Schmitz H, Plessl C. Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study. <i>International Journal of Reconfigurable Computing (IJRC)</i>. 2015;2015. doi:<a href="https://doi.org/10.1155/2015/859425">10.1155/2015/859425</a> T. Kenter, H. Schmitz, and C. Plessl, “Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study,” <i>International Journal of Reconfigurable Computing (IJRC)</i>, vol. 2015, Art. no. 859425, 2015, doi: <a href="https://doi.org/10.1155/2015/859425">10.1155/2015/859425</a>. Kenter, Tobias, Henning Schmitz, and Christian Plessl. “Exploring Tradeoffs between Specialized Kernels and a Reusable Overlay in a Stereo-Matching Case Study.” <i>International Journal of Reconfigurable Computing (IJRC)</i> 2015 (2015). <a href="https://doi.org/10.1155/2015/859425">https://doi.org/10.1155/2015/859425</a>. 2962017-10-17T12:41:49Z2023-09-26T13:29:08Z