{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,25]],"date-time":"2026-06-25T17:24:09Z","timestamp":1782408249716,"version":"3.54.5"},"reference-count":20,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2017,7,19]],"date-time":"2017-07-19T00:00:00Z","timestamp":1500422400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Science Foundation of China","doi-asserted-by":"crossref","award":["U1435219, 61303070, 61402507 and 61402499"],"award-info":[{"award-number":["U1435219, 61303070, 61402507 and 61402499"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Reconfigurable Technol. Syst."],"published-print":{"date-parts":[[2017,9,30]]},"abstract":"<jats:p>Deep convolutional neural networks (CNNs) have gained great success in various computer vision applications. State-of-the-art CNN models for large-scale applications are computation intensive and memory expensive and, hence, are mainly processed on high-performance processors like server CPUs and GPUs. However, there is an increasing demand of high-accuracy or real-time object detection tasks in large-scale clusters or embedded systems, which requires energy-efficient accelerators because of the green computation requirement or the limited battery restriction. Due to the advantages of energy efficiency and reconfigurability, Field-Programmable Gate Arrays (FPGAs) have been widely explored as CNN accelerators. In this article, we present an in-depth analysis of computation complexity and the memory footprint of each CNN layer type. Then a scalable parallel framework is proposed that exploits four levels of parallelism in hardware acceleration. We further put forward a systematic design space exploration methodology to search for the optimal solution that maximizes accelerator throughput under the FPGA constraints such as on-chip memory, computational resources, external memory bandwidth, and clock frequency. Finally, we demonstrate the methodology by optimizing three representative CNNs (LeNet, AlexNet, and VGG-S) on a Xilinx VC709 board. The average performance of the three accelerators is 424.7, 445.6, and 473.4GOP\/s under 100MHz working frequency, which outperforms the CPU and previous work significantly.<\/jats:p>","DOI":"10.1145\/3079758","type":"journal-article","created":{"date-parts":[[2017,7,20]],"date-time":"2017-07-20T17:51:24Z","timestamp":1500573084000},"page":"1-23","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":95,"title":["Throughput-Optimized FPGA Accelerator for Deep Convolutional Neural Networks"],"prefix":"10.1145","volume":"10","author":[{"given":"Zhiqiang","family":"Liu","sequence":"first","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yong","family":"Dou","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingfei","family":"Jiang","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jinwei","family":"Xu","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shijie","family":"Li","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yongmei","family":"Zhou","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingnan","family":"Xu","sequence":"additional","affiliation":[{"name":"National University of Defense Technology, Changsha, Hunan, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2017,7,19]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2014.2339736"},{"key":"e_1_2_1_2_1","doi-asserted-by":"publisher","DOI":"10.1109\/92.784091"},{"key":"e_1_2_1_3_1","volume-title":"Proceedings of the 27th International Conference on Machine Learning (ICML\u201910)","volume":"7","author":"Boureau Lan","year":"2010"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/1815961.1815993"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO.2014.58"},{"key":"e_1_2_1_6_1","doi-asserted-by":"publisher","DOI":"10.1145\/1390156.1390177"},{"key":"e_1_2_1_7_1","first-page":"2148","article-title":"Predicting parameters in deep learning","volume":"7","author":"Denil Misha","year":"2013","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2009.5272559"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.81"},{"key":"e_1_2_1_10_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.223"},{"key":"e_1_2_1_11_1","unstructured":"Alex Krizhevsky Ilya Sutskever and Geoffrey E. Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems. 1097--1105.  Alex Krizhevsky Ilya Sutskever and Geoffrey E. Hinton. 2012. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems. 1097--1105."},{"key":"e_1_2_1_12_1","volume-title":"Proceedings of Twenty-Ninth AAAI Conference on Artificial Intelligence. 2267--2273","author":"Lai Siwei","year":"2015"},{"key":"e_1_2_1_13_1","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847265"},{"key":"e_1_2_1_14_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2014.131"},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1145\/2847263.2847276"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"e_1_2_1_17_1","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2807412"},{"key":"e_1_2_1_18_1","unstructured":"Xilinx. 2015. Virtex7-product-table.pdf. https:\/\/www.xilinx.com\/support\/documentation\/selection-guides.  Xilinx. 2015. Virtex7-product-table.pdf. https:\/\/www.xilinx.com\/support\/documentation\/selection-guides."},{"key":"e_1_2_1_19_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298789"},{"key":"e_1_2_1_20_1","doi-asserted-by":"publisher","DOI":"10.1145\/2684746.2689060"}],"container-title":["ACM Transactions on Reconfigurable Technology and Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3079758","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3079758","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,18]],"date-time":"2025-06-18T03:37:14Z","timestamp":1750217834000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3079758"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2017,7,19]]},"references-count":20,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2017,9,30]]}},"alternative-id":["10.1145\/3079758"],"URL":"https:\/\/doi.org\/10.1145\/3079758","relation":{},"ISSN":["1936-7406","1936-7414"],"issn-type":[{"value":"1936-7406","type":"print"},{"value":"1936-7414","type":"electronic"}],"subject":[],"published":{"date-parts":[[2017,7,19]]},"assertion":[{"value":"2016-06-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2017-03-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2017-07-19","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}