{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T05:34:01Z","timestamp":1790832841434,"version":"4.1.0"},"reference-count":42,"publisher":"Association for Computing Machinery (ACM)","issue":"5s","license":[{"start":{"date-parts":[[2019,10,7]],"date-time":"2019-10-07T00:00:00Z","timestamp":1570406400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["1553419"],"award-info":[{"award-number":["1553419"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Embed. Comput. Syst."],"published-print":{"date-parts":[[2019,10,31]]},"abstract":"<jats:p>\n            This paper addresses the design of systolic array (SA) based convolutional neural network (CNN) accelerators for mobile and embedded domains. On- and off-chip memory accesses to the large activation inputs (sometimes called feature maps) of CNN layers contribute significantly to total energy consumption for such accelerators; while prior has proposed off-chip compression, activations are still stored on-chip in uncompressed form, requiring either large on-chip activation buffers or slow and energy-hungry off-chip accesses. In this paper, we propose CompAct, a new architecture that enables\n            <jats:italic>on-chip<\/jats:italic>\n            compression of activations for SA based CNN accelerators. CompAct is built around several key ideas. First, CompAct identifies an SA schedule that has nearly regular access patterns, enabling the use of a modified run-length coding scheme (RLC). Second, CompAct improves compression ratio of the RLC scheme using Sparse-RLC in later CNN layers and Lossy-RLC in earlier layers. Finally, CompAct proposes look-ahead snoozing that operates synergistically with RLC to reduce the leakage energy of activation buffers. Based on detailed synthesis results, we show that CompAct enables up to 62% reduction in activation buffer energy, and 34% reduction in total chip energy.\n          <\/jats:p>","DOI":"10.1145\/3358178","type":"journal-article","created":{"date-parts":[[2019,10,10]],"date-time":"2019-10-10T13:13:05Z","timestamp":1570713185000},"page":"1-24","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":17,"title":["CompAct"],"prefix":"10.1145","volume":"18","author":[{"given":"Jeff (Jun)","family":"Zhang","sequence":"first","affiliation":[{"name":"New York University, Brooklyn, NY"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Parul","family":"Raj","sequence":"additional","affiliation":[{"name":"New York University, Brooklyn, NY"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shuayb","family":"Zarar","sequence":"additional","affiliation":[{"name":"Microsoft, Redmond, WA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Amol","family":"Ambardekar","sequence":"additional","affiliation":[{"name":"Microsoft, Redmond, WA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Siddharth","family":"Garg","sequence":"additional","affiliation":[{"name":"New York University, Brooklyn, NY"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2019,10,7]]},"reference":[{"key":"e_1_2_1_1_1","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001138"},{"key":"e_1_2_1_2_1","doi-asserted-by":"crossref","unstructured":"Manoj Alwani et al. 2016. Fused-layer CNN accelerators. In IEEE\/ACM MICRO. 1--12.  Manoj Alwani et al. 2016. Fused-layer CNN accelerators. In IEEE\/ACM MICRO. 1--12.","DOI":"10.1109\/MICRO.2016.7783725"},{"key":"e_1_2_1_3_1","unstructured":"ARM. 2018. PROJECT TRILLIUM@ONLINE. https:\/\/www.arm.com\/products\/silicon-ip-cpu\/machine-learning\/project-trillium  ARM. 2018. PROJECT TRILLIUM@ONLINE. https:\/\/www.arm.com\/products\/silicon-ip-cpu\/machine-learning\/project-trillium"},{"key":"e_1_2_1_4_1","doi-asserted-by":"publisher","DOI":"10.1145\/1816038.1815993"},{"key":"e_1_2_1_5_1","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2616357"},{"key":"e_1_2_1_6_1","unstructured":"Reetuparna Das and Tushar Krishna. [n.d.]. DNN Accelerator Architecture -- SIMD or Systolic? https:\/\/www.sigarch.org\/dnn-accelerator-architecture-simd-or-systolic\/. Accessed: 2019-04-27.  Reetuparna Das and Tushar Krishna. [n.d.]. DNN Accelerator Architecture -- SIMD or Systolic? https:\/\/www.sigarch.org\/dnn-accelerator-architecture-simd-or-systolic\/. Accessed: 2019-04-27."},{"key":"e_1_2_1_7_1","volume-title":"Imagenet: A large-scale hierarchical image database","author":"Jia Deng","year":"2009","unstructured":"Jia Deng et al. 2009 . Imagenet: A large-scale hierarchical image database . In IEEE CVPR. 248--255. Jia Deng et al. 2009. Imagenet: A large-scale hierarchical image database. In IEEE CVPR. 248--255."},{"key":"e_1_2_1_8_1","unstructured":"Caiwen Ding Siyu Liao Yanzhi Wang Zhe Li Ning Liu Youwei Zhuo Chao Wang Xuehai Qian Yu Bai Geng Yuan etal 2017. CirCNN: Accelerating and compressing deep neural networks using Block-CirculantWeight matrices. arXiv preprint arXiv:1708.08917 (2017).  Caiwen Ding Siyu Liao Yanzhi Wang Zhe Li Ning Liu Youwei Zhuo Chao Wang Xuehai Qian Yu Bai Geng Yuan et al. 2017. CirCNN: Accelerating and compressing deep neural networks using Block-CirculantWeight matrices. arXiv preprint arXiv:1708.08917 (2017)."},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.1145\/2872887.2750389"},{"key":"e_1_2_1_10_1","doi-asserted-by":"crossref","unstructured":"Li et al. 2018. SmartShuttle: Optimizing off-chip memory accesses for deep learning accelerators. In 2018 IEEE DATE. 343--348.  Li et al. 2018. SmartShuttle: Optimizing off-chip memory accesses for deep learning accelerators. In 2018 IEEE DATE. 343--348.","DOI":"10.23919\/DATE.2018.8342033"},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1145\/3297858.3304014"},{"key":"e_1_2_1_12_1","doi-asserted-by":"crossref","unstructured":"Yijin Guan et al. 2017. FP-DNN: An automated framework for mapping deep neural networks onto FPGAs with RTL-HLS hybrid templates. In 2017 IEEE FCCM. 152--159.  Yijin Guan et al. 2017. FP-DNN: An automated framework for mapping deep neural networks onto FPGAs with RTL-HLS hybrid templates. In 2017 IEEE FCCM. 152--159.","DOI":"10.1109\/FCCM.2017.25"},{"key":"e_1_2_1_13_1","unstructured":"Song Han et al. 2015. Deep compression: Compressing deep neural networks with pruning trained quantization and huffman coding. arXiv preprint arXiv:1510.00149 (2015).  Song Han et al. 2015. Deep compression: Compressing deep neural networks with pruning trained quantization and huffman coding. arXiv preprint arXiv:1510.00149 (2015)."},{"key":"e_1_2_1_14_1","volume-title":"EIE: Efficient inference engine on compressed deep neural network","author":"Song Han","year":"2016","unstructured":"Song Han et al. 2016 . EIE: Efficient inference engine on compressed deep neural network . In Proceedings of IEEE ISCA. 243--254. Song Han et al. 2016. EIE: Efficient inference engine on compressed deep neural network. In Proceedings of IEEE ISCA. 243--254."},{"key":"e_1_2_1_15_1","doi-asserted-by":"publisher","DOI":"10.1166\/jolpe.2018.1575"},{"key":"e_1_2_1_16_1","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2018.2857098"},{"key":"e_1_2_1_17_1","unstructured":"Norman Jouppi et al. 2017. In-datacenter performance analysis of a tensor processing unit. arXiv preprint arXiv:1704.04760 (2017).  Norman Jouppi et al. 2017. In-datacenter performance analysis of a tensor processing unit. arXiv preprint arXiv:1704.04760 (2017)."},{"key":"e_1_2_1_18_1","volume-title":"Proceedings of the IEEE CVPR. 1725--1732","author":"Andrej","unstructured":"Andrej Karpathy et al. 2014. Large-scale video classification with convolutional neural networks . In Proceedings of the IEEE CVPR. 1725--1732 . Andrej Karpathy et al. 2014. Large-scale video classification with convolutional neural networks. In Proceedings of the IEEE CVPR. 1725--1732."},{"key":"e_1_2_1_19_1","unstructured":"Dongyoung Kim et al. 2017. ZeNA: Zero-aware neural network accelerator. IEEE Design Test (2017).  Dongyoung Kim et al. 2017. ZeNA: Zero-aware neural network accelerator. IEEE Design Test (2017)."},{"key":"e_1_2_1_20_1","volume-title":"One weird trick for parallelizing convolutional neural networks. arXiv preprint arXiv:1404.5997","author":"Krizhevsky Alex","year":"2014","unstructured":"Alex Krizhevsky . 2014. One weird trick for parallelizing convolutional neural networks. arXiv preprint arXiv:1404.5997 ( 2014 ). Alex Krizhevsky. 2014. One weird trick for parallelizing convolutional neural networks. arXiv preprint arXiv:1404.5997 (2014)."},{"key":"e_1_2_1_21_1","unstructured":"Alex Krizhevsky et al. 2012. Imagenet classification with deep convolutional neural networks. In NIPS. 1097--1105.  Alex Krizhevsky et al. 2012. Imagenet classification with deep convolutional neural networks. In NIPS. 1097--1105."},{"key":"e_1_2_1_22_1","doi-asserted-by":"crossref","unstructured":"HT Kung etal 2019. Packing sparse convolutional neural networks for efficient systolic array implementations: Column combining under joint optimization. In 24th ASPLOS. ACM 821--834.  HT Kung et al. 2019. Packing sparse convolutional neural networks for efficient systolic array implementations: Column combining under joint optimization. In 24th ASPLOS. ACM 821--834.","DOI":"10.1145\/3297858.3304028"},{"key":"e_1_2_1_23_1","volume-title":"Pruning filters for efficient convnets. arXiv preprint arXiv:1608.08710","author":"Li Hao","year":"2016","unstructured":"Hao Li , Asim Kadav , Igor Durdanovic , Hanan Samet , and Hans Peter Graf . 2016. Pruning filters for efficient convnets. arXiv preprint arXiv:1608.08710 ( 2016 ). Hao Li, Asim Kadav, Igor Durdanovic, Hanan Samet, and Hans Peter Graf. 2016. Pruning filters for efficient convnets. arXiv preprint arXiv:1608.08710 (2016)."},{"key":"e_1_2_1_24_1","volume-title":"Rethinking the value of network pruning. arXiv preprint arXiv:1810.05270","author":"Liu Zhuang","year":"2018","unstructured":"Zhuang Liu , Mingjie Sun , Tinghui Zhou , Gao Huang , and Trevor Darrell . 2018. Rethinking the value of network pruning. arXiv preprint arXiv:1810.05270 ( 2018 ). Zhuang Liu, Mingjie Sun, Tinghui Zhou, Gao Huang, and Trevor Darrell. 2018. Rethinking the value of network pruning. arXiv preprint arXiv:1810.05270 (2018)."},{"key":"e_1_2_1_25_1","volume-title":"Jouppi","author":"Muralimanohar Naveen","year":"2009","unstructured":"Naveen Muralimanohar , Rajeev Balasubramonian , and Norman P . Jouppi . 2009 . CACTI 6.0: A tool to model large caches. HP Laboratories ( 2009), 22--31. Naveen Muralimanohar, Rajeev Balasubramonian, and Norman P. Jouppi. 2009. CACTI 6.0: A tool to model large caches. HP Laboratories (2009), 22--31."},{"key":"e_1_2_1_26_1","first-page":"2","article-title":"SCNN: An accelerator for compressed-sparse convolutional neural networks","volume":"45","author":"Angshuman Parashar","year":"2017","unstructured":"Angshuman Parashar et al. 2017 . SCNN: An accelerator for compressed-sparse convolutional neural networks . SIGARCH Comput. Archit. News 45 , 2 (June 2017). Angshuman Parashar et al. 2017. SCNN: An accelerator for compressed-sparse convolutional neural networks. SIGARCH Comput. Archit. News 45, 2 (June 2017).","journal-title":"SIGARCH Comput. Archit. News"},{"key":"e_1_2_1_27_1","unstructured":"Seongwook Park et al. 2015. 93TOPS\/W scalable deep learning\/inference processor with tetra-parallel MIMD architecture for big-data applications. In IEEE ISSCC.  Seongwook Park et al. 2015. 93TOPS\/W scalable deep learning\/inference processor with tetra-parallel MIMD architecture for big-data applications. In IEEE ISSCC."},{"key":"e_1_2_1_28_1","doi-asserted-by":"crossref","unstructured":"Atul Rahman et al. 2016. Efficient FPGA acceleration of convolutional neural networks using logical-3D compute array. In IEEE DATE. 1393--1398.  Atul Rahman et al. 2016. Efficient FPGA acceleration of convolutional neural networks using logical-3D compute array. In IEEE DATE. 1393--1398.","DOI":"10.3850\/9783981537079_0833"},{"key":"e_1_2_1_29_1","first-page":"805","article-title":"Rotating data for neural network computations","volume":"9","author":"Ross Jonathan","year":"2017","unstructured":"Jonathan Ross and Gregory Michael Thorson . 2017 . Rotating data for neural network computations . US Patent 9 , 805 ,303. Jonathan Ross and Gregory Michael Thorson. 2017. Rotating data for neural network computations. US Patent 9,805,303.","journal-title":"US Patent"},{"key":"e_1_2_1_30_1","volume-title":"IEEE ASAP","author":"Murugan","year":"2009","unstructured":"Murugan Sankaradas et al. 2009. A massively parallel coprocessor for convolutional neural networks . In IEEE ASAP 2009 . 53--60. Murugan Sankaradas et al. 2009. A massively parallel coprocessor for convolutional neural networks. In IEEE ASAP 2009. 53--60."},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2018.2835809"},{"key":"e_1_2_1_32_1","volume-title":"Deep learning in neural networks: An overview. Neural networks 61","author":"Schmidhuber J\u00fcrgen","year":"2015","unstructured":"J\u00fcrgen Schmidhuber . 2015. Deep learning in neural networks: An overview. Neural networks 61 ( 2015 ), 85--117. J\u00fcrgen Schmidhuber. 2015. Deep learning in neural networks: An overview. Neural networks 61 (2015), 85--117."},{"key":"e_1_2_1_33_1","volume-title":"Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556","author":"Simonyan Karen","year":"2014","unstructured":"Karen Simonyan and Andrew Zisserman . 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 ( 2014 ). Karen Simonyan and Andrew Zisserman. 2014. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)."},{"key":"e_1_2_1_34_1","volume-title":"Andrew Anderson, and David Gregg","author":"Aravind Vasudevan","year":"2017","unstructured":"Aravind Vasudevan et al. , Andrew Anderson, and David Gregg . 2017 . Parallel multi channel convolution using general matrix multiplication. In IEEE 28th ASAP. 19--24. Aravind Vasudevan et al., Andrew Anderson, and David Gregg. 2017. Parallel multi channel convolution using general matrix multiplication. In IEEE 28th ASAP. 19--24."},{"key":"e_1_2_1_35_1","doi-asserted-by":"publisher","DOI":"10.1109\/DDECS.2011.5783112"},{"key":"e_1_2_1_36_1","unstructured":"Wei Wen et al. 2016. Learning structured sparsity in deep neural networks. In NIPS. 2074--2082.  Wei Wen et al. 2016. Learning structured sparsity in deep neural networks. In NIPS. 2074--2082."},{"key":"e_1_2_1_37_1","doi-asserted-by":"publisher","DOI":"10.1145\/2964284.2967243"},{"key":"e_1_2_1_38_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.643"},{"key":"e_1_2_1_39_1","volume-title":"Proceedings of ACM ISCA. 548--560","author":"Jiecao","unstructured":"Jiecao Yu et al. 2017. Scalpel: Customizing DNN pruning to the underlying hardware parallelism . In Proceedings of ACM ISCA. 548--560 . Jiecao Yu et al. 2017. Scalpel: Customizing DNN pruning to the underlying hardware parallelism. In Proceedings of ACM ISCA. 548--560."},{"key":"e_1_2_1_40_1","doi-asserted-by":"publisher","DOI":"10.1145\/3195970.3196129"},{"key":"e_1_2_1_41_1","doi-asserted-by":"publisher","DOI":"10.1109\/VTS.2018.8368656"},{"key":"e_1_2_1_42_1","volume-title":"Proceedings of the 2015 ACM FPGA. ACM, 161--170","author":"Chen","unstructured":"Chen Zhang et al. 2015. Optimizing fpga-based accelerator design for deep convolutional neural networks . In Proceedings of the 2015 ACM FPGA. ACM, 161--170 . Chen Zhang et al. 2015. Optimizing fpga-based accelerator design for deep convolutional neural networks. In Proceedings of the 2015 ACM FPGA. ACM, 161--170."}],"container-title":["ACM Transactions on Embedded Computing Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3358178","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3358178","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3358178","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,17]],"date-time":"2025-06-17T22:32:58Z","timestamp":1750199578000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3358178"}},"subtitle":["On-chip &lt;underline&gt;Com&lt;\/underline&gt;pression of &lt;underline&gt;Act&lt;\/underline&gt;ivations for Low Power Systolic Array Based CNN Acceleration"],"short-title":[],"issued":{"date-parts":[[2019,10,7]]},"references-count":42,"journal-issue":{"issue":"5s","published-print":{"date-parts":[[2019,10,31]]}},"alternative-id":["10.1145\/3358178"],"URL":"https:\/\/doi.org\/10.1145\/3358178","relation":{},"ISSN":["1539-9087","1558-3465"],"issn-type":[{"value":"1539-9087","type":"print"},{"value":"1558-3465","type":"electronic"}],"subject":[],"published":{"date-parts":[[2019,10,7]]},"assertion":[{"value":"2019-04-01","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-07-01","order":1,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2019-10-07","order":2,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}