{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,9,20]],"date-time":"2025-09-20T05:42:48Z","timestamp":1758346968530,"version":"3.44.0"},"reference-count":47,"publisher":"Association for Computing Machinery (ACM)","issue":"3","funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"crossref","award":["62025401, 62341407, U24B200170"],"award-info":[{"award-number":["62025401, 62341407, U24B200170"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Beijing Natural Science Foundation","award":["L223004"],"award-info":[{"award-number":["L223004"]}]},{"DOI":"10.13039\/501100013314","name":"\u201c111\u201d project","doi-asserted-by":"crossref","award":["B18001"],"award-info":[{"award-number":["B18001"]}],"id":[{"id":"10.13039\/501100013314","id-type":"DOI","asserted-by":"crossref"}]},{"name":"Damo Academy Research Intern Program"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Archit. Code Optim."],"published-print":{"date-parts":[[2025,9,30]]},"abstract":"<jats:p>\n            Fully homomorphic encryption over torus (TFHE) enables the execution of arbitrary functions on encrypted data through programmable bootstrapping (PBS). However, performing all operations on ciphertext during PBS results in high computational and memory requirements, limiting the deployment of PBS in real-world scenarios. Previous TFHE accelerator designs have attempted to improve performance by employing specific dataflow and functional units, but these techniques may require large off-chip bandwidth or on-chip storage when scaling up computation capacity. Additionally, the design of specialized functional units may limit the utilization of computation units when facing dynamic secure parameter settings. To address these challenges and further improve PBS throughput in TFHE, we propose\n            <jats:sans-serif>Matrix<\/jats:sans-serif>\n            , an ASIC-based architecture that balances off-chip bandwidth and on-chip storage according to the execution flow of PBS. In\n            <jats:sans-serif>Matrix<\/jats:sans-serif>\n            , we utilize a unified special-prime-based processing element (PE) that achieves high utilization with minimal resource overhead. Furthermore, we propose a hybrid PBS dataflow that can efficiently reduce computation complexity and memory requirements. Compared to state-of-the-art TFHE accelerators,\n            <jats:sans-serif>Matrix<\/jats:sans-serif>\n            \u00a0achieves 1.43 \u00d7 -5.66 \u00d7 throughput improvement for PBS. For ZAMA Deep-NN benchmark, we achieve 525.60\u00d7 and 68.06\u00d7 speedup compared to CPU and GPU, respectively.\n            <jats:xref ref-type=\"fn\">\n              <jats:sup>1<\/jats:sup>\n            <\/jats:xref>\n          <\/jats:p>","DOI":"10.1145\/3750446","type":"journal-article","created":{"date-parts":[[2025,7,29]],"date-time":"2025-07-29T11:04:43Z","timestamp":1753787083000},"page":"1-23","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Matrix: Multi-Cipher Structures Dataflow for Parallel and Pipelined TFHE Accelerator"],"prefix":"10.1145","volume":"22","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8534-6494","authenticated-orcid":false,"given":"Ling","family":"Liang","sequence":"first","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7067-492X","authenticated-orcid":false,"given":"Zhen","family":"Gu","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-8613-5323","authenticated-orcid":false,"given":"Fahong","family":"Zhang","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6468-7575","authenticated-orcid":false,"given":"Zhaohui","family":"Chen","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-6513-5453","authenticated-orcid":false,"given":"Zhirui","family":"Li","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-4658-8192","authenticated-orcid":false,"given":"Xin","family":"Fan","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8440-3875","authenticated-orcid":false,"given":"Dimin","family":"Niu","sequence":"additional","affiliation":[{"name":"Alibaba Group","place":["Sunnyvale, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7212-2264","authenticated-orcid":false,"given":"Meng","family":"Li","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0436-6670","authenticated-orcid":false,"given":"Zhiyong","family":"Li","sequence":"additional","affiliation":[{"name":"Alibaba DAMO Academy","place":["Beijing, China"]},{"name":"KAIST","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6297-2700","authenticated-orcid":false,"given":"Zongwei","family":"Wang","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7696-9799","authenticated-orcid":false,"given":"Hongzhong","family":"Zheng","sequence":"additional","affiliation":[{"name":"Alibaba Group","place":["Sunnyvale, United States"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6854-8211","authenticated-orcid":false,"given":"Yimao","family":"Cai","sequence":"additional","affiliation":[{"name":"Peking University","place":["Beijing, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2093-1788","authenticated-orcid":false,"given":"Yuan","family":"Xie","sequence":"additional","affiliation":[{"name":"The Hong Kong University of Science and Technology","place":["Hong Kong, China"]}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,9,19]]},"reference":[{"key":"e_1_3_2_2_2","doi-asserted-by":"publisher","DOI":"10.1007\/s00145-019-09319-x"},{"key":"e_1_3_2_3_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-32009-5_50"},{"key":"e_1_3_2_4_2","article-title":"Somewhat practical fully homomorphic encryption","author":"Fan Junfeng","year":"2012","unstructured":"Junfeng Fan and Frederik Vercauteren. 2012. Somewhat practical fully homomorphic encryption. Cryptology ePrint Archive (2012), 144.","journal-title":"Cryptology ePrint Archive"},{"key":"e_1_3_2_5_2","doi-asserted-by":"publisher","DOI":"10.1145\/2633600"},{"key":"e_1_3_2_6_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-70694-8_15"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-78086-9_1"},{"key":"e_1_3_2_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS57527.2023.00012"},{"key":"e_1_3_2_9_2","first-page":"532","volume-title":"Proceedings of the 2022 IEEE 15th International Conference on Cloud Computing.","author":"Carpov Sergiu","year":"2022","unstructured":"Sergiu Carpov, Nicolas Gama, Mariya Georgieva, and Dimitar Jetchev. 2022. GenoPPML\u2013a framework for genomic privacy-preserving machine learning. In Proceedings of the 2022 IEEE 15th International Conference on Cloud Computing. IEEE, 532\u2013542."},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/3548606.3560702"},{"key":"e_1_3_2_11_2","article-title":"At last! a homomorphic aes evaluation in less than 30 seconds by means of TFHE","author":"Trama Daphn\u00e9","year":"2023","unstructured":"Daphn\u00e9 Trama, Pierre-Emmanuel Clet, Aymen Boudguiga, and Renaud Sirdey. 2023. At last! a homomorphic aes evaluation in less than 30 seconds by means of TFHE. Cryptology ePrint Archive (2023), 1020.","journal-title":"Cryptology ePrint Archive"},{"key":"e_1_3_2_12_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-22966-4_5"},{"key":"e_1_3_2_13_2","article-title":"Efficient TFHE bootstrapping in the multiparty setting","author":"Park Jeongeun","year":"2023","unstructured":"Jeongeun Park and Sergi Rovira. 2023. Efficient TFHE bootstrapping in the multiparty setting. Cryptology ePrint Archive 11 (2023), 118625\u2013118638.","journal-title":"Cryptology ePrint Archive"},{"key":"e_1_3_2_14_2","doi-asserted-by":"publisher","DOI":"10.1145\/3319535.3354226"},{"key":"e_1_3_2_15_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33019466"},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3159694"},{"key":"e_1_3_2_17_2","doi-asserted-by":"crossref","unstructured":"A. Putra Prasetiyo Y. Chen J. Kim and J. Kim. 2023. Strix: An end-to-end streaming architecture with two-level ciphertext batching for fully homomorphic encryption with programmable bootstrapping. 1319\u20131331.","DOI":"10.1145\/3613424.3614264"},{"key":"e_1_3_2_18_2","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530435"},{"key":"e_1_3_2_19_2","doi-asserted-by":"publisher","DOI":"10.1145\/3569955"},{"key":"e_1_3_2_20_2","unstructured":"N. Team. 2020. Nufhe: A gpu-accelerated fully homomorphic encryption library. https:\/\/github.com\/nucypher\/nufhe. Last accessed on 2025-1-10."},{"key":"e_1_3_2_21_2","doi-asserted-by":"crossref","unstructured":"M. V. Beirendonck J. D\u2019Anvers F. Turan and I. Verbauwhede. 2023. FPT: A fixed-point accelerator for torus fully homomorphic encryption. 741\u2013755.","DOI":"10.1145\/3576915.3623159"},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","unstructured":"A. Putra Prasetiyo and J. Kim. 2024. Morphling: A throughput-maximized TFHE-based accelerator using transform-domain reuse. In IEEE International Symposium on High-Performance Computer Architecture HPCA 2024 Edinburgh United Kingdom March 2-6 2024. IEEE 249\u2013262.","DOI":"10.1109\/HPCA57654.2024.00028"},{"key":"e_1_3_2_23_2","unstructured":"K. Matsuoka. 2020. TFHEpp: pure C++ implementation of TFHE cryptosystem. https:\/\/github.com\/virtualsecureplatform\/TFHEpp. Last accessed on 2025-1-25."},{"key":"e_1_3_2_24_2","article-title":"MOSFHET: Optimized software for FHE over the torus","author":"Guimar\u00e3es Antonio","year":"2024","unstructured":"Antonio Guimar\u00e3es, Edson Borin, and Diego F. Aranha. 2024. MOSFHET: Optimized software for FHE over the torus. Cryptology ePrint Archive 14, 3 (2024), 577\u2013593.","journal-title":"Cryptology ePrint Archive"},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.array.2021.100118"},{"key":"e_1_3_2_26_2","volume-title":"Proceedings of the WAHC 2020-8th Workshop on Encrypted Computing and Applied Homomorphic Cryptography","author":"Chillotti Ilaria","year":"2020","unstructured":"Ilaria Chillotti, Marc Joye, Damien Ligier, Jean-Baptiste Orfila, and Samuel Tap. 2020. CONCRETE: Concrete operates on ciphertexts rapidly by extending TfhE. In Proceedings of the WAHC 2020-8th Workshop on Encrypted Computing and Applied Homomorphic Cryptography."},{"key":"e_1_3_2_27_2","doi-asserted-by":"crossref","unstructured":"N. Samardzic A. Feldmann A. Krastev S. Devadas R. G. Dreslinski C. Peikert and D. S\u00e1nchez. 2021. F1: A fast and programmable accelerator for fully homomorphic encryption. 238\u2013252.","DOI":"10.1145\/3466752.3480070"},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00086"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527415"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1109\/TASSP.1975.1162687"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAU.1969.1162035"},{"key":"e_1_3_2_32_2","doi-asserted-by":"publisher","DOI":"10.1109\/TAU.1969.1162049"},{"key":"e_1_3_2_33_2","doi-asserted-by":"publisher","DOI":"10.1137\/S1064827593247023"},{"key":"e_1_3_2_34_2","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2867655"},{"key":"e_1_3_2_35_2","doi-asserted-by":"publisher","DOI":"10.1145\/3508352.3549413"},{"key":"e_1_3_2_36_2","doi-asserted-by":"publisher","DOI":"10.14778\/3574245.3574248"},{"key":"e_1_3_2_37_2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-38551-3_2"},{"volume-title":"Proceedings of the 60th ACM\/IEEE Design Automation Conference, San Francisco, California, USA, July 9-13, 2023","year":"2023","key":"e_1_3_2_38_2","unstructured":"Xuanle, Zhaohui Chen, Zhen Gu, Yanheng Lu, Ruiguang Zhong, Wen jie Lu, Jiansong Zhang, Yichi Zhang, Hanghang Wu, Xiaofu Zheng, et\u00a0al. 2023. CHAM: A customized homomorphic encryption accelerator for fast matrix-vector product. In Proceedings of the 60th ACM\/IEEE Design Automation Conference, San Francisco, California, USA, July 9-13, 2023. ACM."},{"key":"e_1_3_2_39_2","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247836"},{"key":"e_1_3_2_40_2","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3657375"},{"key":"e_1_3_2_41_2","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2010.93"},{"key":"e_1_3_2_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.3017930"},{"key":"e_1_3_2_43_2","doi-asserted-by":"publisher","DOI":"10.1109\/IMW.2017.7939084"},{"key":"e_1_3_2_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2016.2602221"},{"key":"e_1_3_2_45_2","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527393"},{"key":"e_1_3_2_46_2","doi-asserted-by":"publisher","DOI":"10.1145\/3576915.3616608"},{"key":"e_1_3_2_47_2","doi-asserted-by":"publisher","DOI":"10.1145\/2732516.2732520"},{"key":"e_1_3_2_48_2","article-title":"E3: A framework for compiling C++ programs with encrypted operands","author":"Chielle Eduardo","year":"2018","unstructured":"Eduardo Chielle, Oleg Mazonka, Homer Gamil, Nektarios Georgios Tsoutsos, and Michail Maniatakos. 2018. E3: A framework for compiling C++ programs with encrypted operands. Cryptology ePrint Archive (2018), 1013.","journal-title":"Cryptology ePrint Archive"}],"container-title":["ACM Transactions on Architecture and Code Optimization"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3750446","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,20]],"date-time":"2025-09-20T00:49:28Z","timestamp":1758329368000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3750446"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,9,19]]},"references-count":47,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2025,9,30]]}},"alternative-id":["10.1145\/3750446"],"URL":"https:\/\/doi.org\/10.1145\/3750446","relation":{},"ISSN":["1544-3566","1544-3973"],"issn-type":[{"type":"print","value":"1544-3566"},{"type":"electronic","value":"1544-3973"}],"subject":[],"published":{"date-parts":[[2025,9,19]]},"assertion":[{"value":"2024-12-16","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-07-14","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-09-19","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}