{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T17:34:51Z","timestamp":1786642491171,"version":"build-2736575974"},"reference-count":156,"publisher":"Association for Computing Machinery (ACM)","issue":"3","license":[{"start":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T00:00:00Z","timestamp":1786579200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"DOI":"10.13039\/501100003977","name":"Israel Science Foundation","doi-asserted-by":"publisher","award":["1027\/18, and 1998\/22"],"award-info":[{"award-number":["1027\/18, and 1998\/22"]}],"id":[{"id":"10.13039\/501100003977","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100024250","name":"Israel Innovation Authority","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100024250","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100016296","name":"Israel National Cyber Directorate","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100016296","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100018707","name":"Technion Hiroshi Fujiwara Cyber Security Research Center","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100018707","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Comput. Syst."],"published-print":{"date-parts":[[2026,8,31]]},"abstract":"<jats:p>We propose a new system design for connecting hardware and FPGA accelerators to the network, allowing them to directly control commodity ASIC NICs without using the CPU. This solves the key challenge of leveraging existing NIC hardware offloads such as RDMA and virtualization for hardware disaggregation and accelerator networking. Our approach supports a diverse set of use cases, from direct network access for disaggregated accelerators to inline acceleration of the network stack and disaggregation of main system memory, all without implementing complex networking logic.<\/jats:p>\n                  <jats:p>\n                    To demonstrate this approach, we build\n                    <jats:bold>FlexDriver (FLD)<\/jats:bold>\n                    , a hardware module that implements a NIC data-plane driver. Our main technical contribution is compressing NIC control structures by\n                    <jats:inline-formula content-type=\"math\/tex\">\n                      <jats:tex-math notation=\"LaTeX\" version=\"MathJax\">\\(5\\times\\)<\/jats:tex-math>\n                    <\/jats:inline-formula>\n                    , allowing FLD to achieve high scalability with low die area and no memory bandwidth interference.\n                  <\/jats:p>\n                  <jats:p>We build two prototypes \u2013 FLD core on NVIDIA Innova-2 FPGA SmartNICs and FLD with a load\/store interface on IBM OpenCAPI FPGA deployment with ConnectX-6 Dx. We demonstrate four different use cases: a disaggregated LTE cipher, an IP-reassembly inline accelerator, an IoT cryptographic-token authentication offload, and a fine-grained memory disaggregation datapath over RDMA. These leverage the ASIC NIC for RDMA processing, VXLAN tunneling, and traffic shaping, without CPU involvement.<\/jats:p>","DOI":"10.1145\/3822399","type":"journal-article","created":{"date-parts":[[2026,6,24]],"date-time":"2026-06-24T21:05:11Z","timestamp":1782335111000},"page":"1-35","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["FlexDriver: A Network Driver for Your Accelerator"],"prefix":"10.1145","volume":"44","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2159-9046","authenticated-orcid":false,"given":"Haggai","family":"Eran","sequence":"first","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]},{"name":"Technion Israel Institute of Technology","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2344-5665","authenticated-orcid":false,"given":"Dimitris","family":"Syrivelis","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Athens, Greece"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-7360-5617","authenticated-orcid":false,"given":"Maxim","family":"Fudim","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-1509-637X","authenticated-orcid":false,"given":"Gabi","family":"Malka","sequence":"additional","affiliation":[{"name":"Technion Israel Institute of Technology","place":["Haifa, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1636-7531","authenticated-orcid":false,"given":"Gal","family":"Shalom","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]},{"name":"Technion Israel Institute of Technology","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-2771-1345","authenticated-orcid":false,"given":"Noam","family":"Cohen","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-5542-4064","authenticated-orcid":false,"given":"Amit","family":"Hermony","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5561-9069","authenticated-orcid":false,"given":"Dotan","family":"Levi","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2661-7906","authenticated-orcid":false,"given":"Liran","family":"Liss","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3635-1641","authenticated-orcid":false,"given":"Paraskevas","family":"Bakopoulos","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Athens, Greece"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0004-0183-1612","authenticated-orcid":false,"given":"Elad","family":"Mentovich","sequence":"additional","affiliation":[{"name":"NVIDIA Corp","place":["Yokneam Ilit, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9659-068X","authenticated-orcid":false,"given":"Mark","family":"Silberstein","sequence":"additional","affiliation":[{"name":"Electrical Engineering, Technion Israel Institute of Technology Faculty of Electrical Engineering","place":["Haifa, Israel"]},{"name":"NVIDIA Corp","place":["Haifa, Israel"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"320","published-online":{"date-parts":[[2026,8,13]]},"reference":[{"key":"e_1_3_1_2_2","doi-asserted-by":"publisher","DOI":"10.1109\/HOTI.2017.13"},{"key":"e_1_3_1_3_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2017.12.007"},{"key":"e_1_3_1_4_2","unstructured":"Alpha Data Ltd. 2023. (Alpha Data 9H3 Platform). Retrieved July 2026 from https:\/\/www.alpha-data.com\/product\/adm-pcie-9h3\/"},{"key":"e_1_3_1_5_2","unstructured":"Amazon Web Services. 2019. AWS Nitro System. Retrieved July 2026 from https:\/\/aws.amazon.com\/ec2\/nitro\/"},{"key":"e_1_3_1_6_2","unstructured":"John Appleby. 2014. SAP HANA \u2013 Scale-up or Scale-out Hardware? Retrieved July 2026 from https:\/\/blogs.sap.com\/2014\/12\/10\/sap-hana-scale-scale-hardware\/"},{"key":"e_1_3_1_7_2","volume-title":"Proceedings of the 2015 Storage Developer Conference","author":"Bates Stephen","year":"2015","unstructured":"Stephen Bates. 2015. Donard: NVM express for Peer-2-Peer between SSDs and other PCIe devices. In Proceedings of the 2015 Storage Developer Conference. SNIA, Santa Clara, CA, USA."},{"key":"e_1_3_1_8_2","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2015.12"},{"key":"e_1_3_1_9_2","doi-asserted-by":"publisher","DOI":"10.1109\/DRCN48652.2020.1570611009"},{"key":"e_1_3_1_10_2","doi-asserted-by":"publisher","DOI":"10.1145\/1879141.1879175"},{"key":"e_1_3_1_11_2","doi-asserted-by":"publisher","DOI":"10.1145\/3309987"},{"key":"e_1_3_1_12_2","unstructured":"Broadcom. 2019. NVMe over Fabrics Performance Stingray\u2122-Based Storage Appliance. Retrieved July 2026 from https:\/\/docs.broadcom.com\/doc\/broadcom-stingray-100G-NVMe-oF-performance"},{"key":"e_1_3_1_13_2","unstructured":"Broadcom. 2020. NetXtreme E-Series PCIe NIC Ethernet Adapters Specification Sheet. Retrieved July 2026 from https:\/\/docs.broadcom.com\/doc\/NetXtreme-E-PCIENIC-SG"},{"key":"e_1_3_1_14_2","volume-title":"Proceedings of the OSDI\u201920","author":"Brunella Marco Spaziani","year":"2020","unstructured":"Marco Spaziani Brunella, Giacomo Belocchi, Marco Bonola, Salvatore Pontarelli, Giuseppe Siracusano, Giuseppe Bianchi, Aniello Cammarano, Alessandro Palumbo, Luca Petrucci, and Roberto Bifulco. 2020. hXDP: Efficient software packet processing on FPGA NICs. In Proceedings of the OSDI\u201920. USENIX Association."},{"key":"e_1_3_1_15_2","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446713"},{"key":"e_1_3_1_16_2","doi-asserted-by":"publisher","DOI":"10.5555\/3195638.3195647"},{"key":"e_1_3_1_17_2","unstructured":"Chelsio Communications. 2018. Accelerated IPsec-VPN Communication with T6. Retrieved July 2026 from https:\/\/www.chelsio.com\/wp-content\/uploads\/resources\/t6-100g-ipsec-linux.pdf"},{"key":"e_1_3_1_18_2","doi-asserted-by":"publisher","DOI":"10.1145\/3241793.3241795"},{"key":"e_1_3_1_19_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2018.022071131"},{"key":"e_1_3_1_20_2","unstructured":"Cisco. 2020. Pre-Fragmentation for IPsec VPNs. Retrieved July 2026 from https:\/\/www.cisco.com\/c\/en\/us\/td\/docs\/ios-xml\/ios\/sec_conn_dplane\/configuration\/xe-16-12\/sec-ipsec-data-plane-xe-16-12-book\/sec-pre-frag-vpns.pdf"},{"key":"e_1_3_1_21_2","unstructured":"NVIDIA Corporation. 2020. NVIDIA Multi-Instance GPU. Retrieved July 2026 from https:\/\/www.nvidia.com\/en-us\/technologies\/multi-instance-gpu\/"},{"key":"e_1_3_1_22_2","unstructured":"NVIDIA Corporation. 2024. NVIDIA Holoscan Sensor Bridge. Retrieved July 2026 from https:\/\/www.nvidia.com\/en-eu\/technologies\/holoscan-sensor-bridge\/"},{"key":"e_1_3_1_23_2","unstructured":"Al Danial. 2021. cloc\u2013count lines of code. Retrieved July 2026 from https:\/\/github.com\/AlDanial\/cloc"},{"key":"e_1_3_1_24_2","doi-asserted-by":"publisher","DOI":"10.1145\/2931088.2931091"},{"key":"e_1_3_1_25_2","unstructured":"Sujal Das and Abhijeet Prabhune. 2016. Netronome \u2013 The Case for Express Virtio (XVIO). Retrieved July 2026 from https:\/\/www.netronome.com\/blog\/the-case-for-express-virtio-xvio-part-1\/"},{"key":"e_1_3_1_26_2","doi-asserted-by":"publisher","DOI":"10.1145\/3669900"},{"key":"e_1_3_1_27_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jpdc.2012.01.020"},{"key":"e_1_3_1_28_2","unstructured":"DPDK. 2018. Generic flow API (rte_flow). Retrieved July 2026 from https:\/\/doc.dpdk.org\/guides\/prog_guide\/ethdev\/flow_offload.html"},{"key":"e_1_3_1_29_2","unstructured":"DPDK. 2019. MLX5 Poll Mode Driver. Retrieved July 2026 from https:\/\/doc.dpdk.org\/guides\/nics\/mlx5.html"},{"key":"e_1_3_1_30_2","unstructured":"Haggai Eran. 2021. FlexDriver Performance and Memory Utilization Models. Retrieved from https:\/\/github.com\/acsl-technion\/flexdriver-model"},{"key":"e_1_3_1_31_2","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507776"},{"key":"e_1_3_1_32_2","first-page":"345","volume-title":"Proceedings of the USENIX ATC 19","author":"Eran Haggai","year":"2019","unstructured":"Haggai Eran, Lior Zeno, Maroun Tork, Gabi Malka, and Mark Silberstein. 2019. NICA: An infrastructure for inline acceleration of network applications. In Proceedings of the USENIX ATC 19. USENIX Association, Renton, WA, 345\u2013362."},{"key":"e_1_3_1_33_2","unstructured":"Mesut Ergin Harry Van Haaren and Charlie Tai. 2019. Intel Ethernet Controller 700 Series - Open vSwitch Hardware Acceleration Application Note. Retrieved from https:\/\/builders.intel.com\/docs\/networkbuilders\/intel-ethernet-controller-700-series-open-vswitch-hardware-acceleration-application-note.pdf"},{"key":"e_1_3_1_34_2","doi-asserted-by":"publisher","DOI":"10.1145\/3289602.3293909"},{"key":"e_1_3_1_35_2","first-page":"673","volume-title":"Proceedings of the USENIX ATC 20","author":"Farshin Alireza","year":"2020","unstructured":"Alireza Farshin, Amir Roozbeh, Gerald Q. Maguire Jr., and Dejan Kosti\u0107. 2020. Reexamining direct cache access to optimize I\/O intensive applications for multi-hundred-gigabit networks. In Proceedings of the USENIX ATC 20. USENIX Association, 673\u2013689."},{"key":"e_1_3_1_36_2","volume-title":"Proceedings of the NSDI 18","author":"Firestone Daniel","year":"2018","unstructured":"Daniel Firestone, Andrew Putnam, Sambhrama Mundkur, Derek Chiou, Alireza Dabagh, Mike Andrewartha, Hari Angepat, Vivek Bhanu, Adrian Caulfield, Eric Chung, et\u00a0al. 2018. Azure accelerated networking: SmartNICs in the public cloud. In Proceedings of the NSDI 18. USENIX Association, Renton, WA."},{"key":"e_1_3_1_37_2","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM48280.2020.00015"},{"key":"e_1_3_1_38_2","unstructured":"Oded Gabbay. 2020. Adding GAUDI NIC code to habanalabs driver. Retrieved July 2026 from https:\/\/lwn.net\/Articles\/831227\/"},{"key":"e_1_3_1_39_2","unstructured":"Gen-Z Consortium. 2016. Gen-Z Overview. Retrieved July 2026 from https:\/\/genzconsortium.org\/wp-content\/uploads\/2018\/05\/Gen-Z-Overview-V1.pdf"},{"key":"e_1_3_1_40_2","unstructured":"Google Cloud. 2020. Google Cloud VPN Documentation \u2013 MTU Considerations. Retrieved July 2026 from https:\/\/cloud.google.com\/network-connectivity\/docs\/vpn\/concepts\/mtu-considerations"},{"key":"e_1_3_1_41_2","first-page":"287","volume-title":"Proceedings of the USENIX ATC 22","author":"Gouk Donghyun","year":"2022","unstructured":"Donghyun Gouk, Sangwon Lee, Miryeong Kwon, and Myoungsoo Jung. 2022. Direct access, High-performance memory disaggregation with DirectCXL. In Proceedings of the USENIX ATC 22. USENIX Association, Carlsbad, CA, 287\u2013294."},{"key":"e_1_3_1_42_2","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2009.33"},{"key":"e_1_3_1_43_2","first-page":"649","volume-title":"Proceedings of the NSDI 17.","author":"Gu Juncheng","year":"2017","unstructured":"Juncheng Gu, Youngmoon Lee, Yiwen Zhang, Mosharaf Chowdhury, and Kang G. Shin. 2017. Efficient memory disaggregation with INFINISWAP. In Proceedings of the NSDI 17.USENIX Association, USA, 649\u2013667."},{"key":"e_1_3_1_44_2","doi-asserted-by":"publisher","DOI":"10.1109\/CLOUD.2019.00064"},{"key":"e_1_3_1_45_2","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507762"},{"key":"e_1_3_1_46_2","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126970"},{"key":"e_1_3_1_47_2","unstructured":"Bert Hubert Thomas Graf Gregory Maxwell Remco van Mook Martijn van Oosterhout Paul B. Schroeder Jasper Spaans and Pedro Larroy. 2012. Linux Advanced Routing and Traffic Control HOWTO. Retrieved from https:\/\/lartc.org\/howto\/"},{"key":"e_1_3_1_48_2","volume-title":"InfiniBand Architecture Specification \u2013 Release 1.2","author":"Association InfiniBand Trade","year":"2004","unstructured":"InfiniBand Trade Association. 2004. Shared Receive Queue. In InfiniBand Architecture Specification \u2013 Release 1.2. Vol. 1. Chapter 10.2.9."},{"key":"e_1_3_1_49_2","unstructured":"Intel. 2013. Intel Ethernet Converged Network Adapter X540. Retrieved July 2026 from https:\/\/www.intel.com\/content\/dam\/www\/public\/us\/en\/documents\/product-briefs\/ethernet-x540-t2-brief.pdf"},{"key":"e_1_3_1_50_2","article-title":"Intel QuickAssist Adapter 8950 Product Brief","year":"2015","unstructured":"Intel. 2015. Intel QuickAssist Adapter 8950 Product Brief. Retrieved July 2026 from https:\/\/www.intel.com\/content\/dam\/www\/public\/us\/en\/documents\/product-briefs\/quickassist-adapter-8950-brief.pdf","journal-title":"R"},{"key":"e_1_3_1_51_2","unstructured":"Intel. 2018. Intel RackScale Architecture. Retrieved July 2026 from https:\/\/www.intel.com\/content\/www\/us\/en\/architecture-and-technology\/rack-scale-design\/rack-scale-design-architecture-white-paper.html"},{"key":"e_1_3_1_52_2","unstructured":"Intel. 2019. Intel 82599 10 GbE Controller Datasheet. Retrieved July 2026 from https:\/\/www.intel.com\/content\/www\/us\/en\/content-details\/331520\/intel-82599-10-gigabit-ethernet-controller-datasheet.html"},{"key":"e_1_3_1_53_2","unstructured":"Intel. 2019. Intel\u00ae Multi-Buffer Crypto for IPsec Library. Retrieved July 2026 from https:\/\/github.com\/intel\/intel-ipsec-mb"},{"key":"e_1_3_1_54_2","unstructured":"Intel. 2020. Intel Programmable Acceleration Card (PAC) with Intel Arria 10 GX FPGA Data Sheet. Retrieved July 2026 from https:\/\/www.intel.com\/content\/www\/us\/en\/docs\/programmable\/683226\/current\/introduction-rush-creek.html"},{"key":"e_1_3_1_55_2","unstructured":"Intel. 2020. Intel Scalable I\/O Virtualization Technical Specifications. Retrieved July 2026 from https:\/\/software.intel.com\/content\/www\/us\/en\/develop\/download\/intel-scalable-io-virtualization-technical-specification.html"},{"key":"e_1_3_1_56_2","unstructured":"Intel. 2021. Intel FPGA Programmable Acceleration Card N3000 Data Sheet. Retrieved July 2026 from https:\/\/www.intel.com\/content\/www\/us\/en\/docs\/programmable\/683260\/1-1\/introduction-vista-creek.html"},{"key":"e_1_3_1_57_2","doi-asserted-by":"publisher","DOI":"10.14778\/3137628.3137632"},{"key":"e_1_3_1_58_2","volume-title":"Proceedings of the NSDI 16","author":"Istv\u00e1n Zsolt","year":"2016","unstructured":"Zsolt Istv\u00e1n, David Sidler, Gustavo Alonso, and Marko Vukolic. 2016. Consensus in a box: Inexpensive coordination in hardware. In Proceedings of the NSDI 16. USENIX Association, Santa Clara, CA."},{"key":"e_1_3_1_59_2","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2018.00029"},{"key":"e_1_3_1_60_2","volume-title":"PCI Express Technology: Comprehensive Guide to Generations 1.x, 2.x, 3.0","author":"Jackson Mike","year":"2012","unstructured":"Mike Jackson and Ravi Budruk. 2012. PCI Express Technology: Comprehensive Guide to Generations 1.x, 2.x, 3.0. MindShare, Inc."},{"key":"e_1_3_1_61_2","unstructured":"Jens Axboe. 2017. fio - Flexible I\/O tester. Retrieved July 2026 from https:\/\/fio.readthedocs.io\/en\/latest\/fio_doc.html"},{"key":"e_1_3_1_62_2","doi-asserted-by":"publisher","DOI":"10.17487\/RFC7519"},{"key":"e_1_3_1_63_2","volume-title":"Proceedings of the NSDI 19","author":"Kalia Anuj","year":"2019","unstructured":"Anuj Kalia, Michael Kaminsky, and David Andersen. 2019. Datacenter RPCs can be general and fast. In Proceedings of the NSDI 19. USENIX Association, Boston, MA."},{"key":"e_1_3_1_64_2","doi-asserted-by":"publisher","DOI":"10.1145\/2619239.2626299"},{"key":"e_1_3_1_65_2","volume-title":"Proceedings of the USENIX ATC 16","author":"Kalia Anuj","year":"2016","unstructured":"Anuj Kalia, Michael Kaminsky, and David G. Andersen. 2016. Design guidelines for high performance RDMA systems. In Proceedings of the USENIX ATC 16. USENIX Association, Denver, CO."},{"key":"e_1_3_1_66_2","doi-asserted-by":"publisher","DOI":"10.1145\/2872362.2872367"},{"key":"e_1_3_1_67_2","first-page":"715","volume-title":"Proceedings of the USENIX ATC 21","author":"Kim Jonghyeon","year":"2021","unstructured":"Jonghyeon Kim, Wonkyo Choe, and Jeongseob Ahn. 2021. Exploring the design space of page management for Multi-tiered memory systems. In Proceedings of the USENIX ATC 21. USENIX Association, 715\u2013728. Retrieved from https:\/\/www.usenix.org\/conference\/atc21\/presentation\/kim-jonghyeon"},{"key":"e_1_3_1_68_2","volume-title":"Proceedings of the OSDI\u201914","author":"Kim Sangman","year":"2014","unstructured":"Sangman Kim, Seonggu Huh, Xinya Zhang, Yige Hu, Amir Wated, Emmett Witchel, and Mark Silberstein. 2014. GPUnet: Networking abstractions for GPU programs. In Proceedings of the OSDI\u201914. USENIX Association, Broomfield, CO."},{"key":"e_1_3_1_69_2","volume-title":"Proceedings of the OSDI\u201920","author":"Korolija Dario","year":"2020","unstructured":"Dario Korolija, Timothy Roscoe, and Gustavo Alonso. 2020. Do OS abstractions make sense on FPGAs?. In Proceedings of the OSDI\u201920. USENIX Association."},{"key":"e_1_3_1_70_2","unstructured":"Patrick Kutch. 2011. PCI-SIG SR-IOV primer: An introduction to SR-IOV technology. Retrieved from http:\/\/www.intel.com\/content\/dam\/doc\/application-note\/pci-sig-sr-iov-primer-sr-iov-technology-paper.pdfIntel application note 321211\u2013002."},{"key":"e_1_3_1_71_2","doi-asserted-by":"publisher","DOI":"10.1145\/3126908.3126950"},{"key":"e_1_3_1_72_2","doi-asserted-by":"publisher","DOI":"10.1145\/2934872.2934897"},{"key":"e_1_3_1_73_2","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3578835"},{"key":"e_1_3_1_74_2","doi-asserted-by":"publisher","DOI":"10.1145\/1555815.1555789"},{"key":"e_1_3_1_75_2","first-page":"243","volume-title":"Proceedings of the OSDI\u201920","author":"Lin Jiaxin","year":"2020","unstructured":"Jiaxin Lin, Kiran Patel, Brent E. Stephens, Anirudh Sivaraman, and Aditya Akella. 2020. PANIC: A high-performance programmable NIC for multi-tenant networks. In Proceedings of the OSDI\u201920. USENIX Association, 243\u2013259."},{"key":"e_1_3_1_76_2","unstructured":"Linux Kernel Documentation. 2007. Memory Hot(Un)Plug. Retrieved July 2026 from https:\/\/docs.kernel.org\/admin-guide\/mm\/memory-hotplug.html"},{"key":"e_1_3_1_77_2","unstructured":"Linux Kernel Documentation. 2015. Direct Access for Files (DAX). Retrieved July 2026 from https:\/\/docs.kernel.org\/filesystems\/dax.html"},{"key":"e_1_3_1_78_2","unstructured":"Liran Liss. 2018. NVMf Target Offload. Retrieved July 2026 from https:\/\/www.openfabrics.org\/images\/2018workshop\/presentations\/308_LLiss_NVMfTargetOffload.pdf"},{"key":"e_1_3_1_79_2","doi-asserted-by":"publisher","DOI":"10.23919\/USNC-URSINRSM66067.2025.10906844"},{"key":"e_1_3_1_80_2","doi-asserted-by":"publisher","DOI":"10.1145\/1330107.1330129"},{"key":"e_1_3_1_81_2","unstructured":"Layong Larry Luo. 2018. Towards Converged SmartNIC Architecture for Bare Metal and Public Clouds. Retrieved July 2026 from https:\/\/conferences.sigcomm.org\/events\/apnet2018\/slides\/larry.pdf"},{"key":"e_1_3_1_82_2","unstructured":"Gabi Malka. 2021. FlexDriver IoT authentication offload example AFU. Retrieved from https:\/\/github.com\/acsl-technion\/flexdriver-iot-auth"},{"key":"e_1_3_1_83_2","unstructured":"Gabi Malka. 2021. FlexDriver ZUC cipher example AFU. Retrieved from https:\/\/github.com\/acsl-technion\/flexdriver-zuc"},{"key":"e_1_3_1_84_2","first-page":"19","article-title":"Memory bandwidth and machine balance in current high performance computers","author":"McCalpin John D.","year":"1995","unstructured":"John D. McCalpin. 1995. Memory bandwidth and machine balance in current high performance computers. IEEE Computer Society Technical Committee on Computer Architecture Newsletter(1995), 19\u201325.","journal-title":"IEEE Computer Society Technical Committee on Computer Architecture Newsletter"},{"key":"e_1_3_1_85_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2020.2975185"},{"key":"e_1_3_1_86_2","unstructured":"Mellanox Technologies. 2017. Innova\u2122 Flex 4 Lx EN Adapter Card Product Brief. Retrieved July 2026 from https:\/\/www.mellanox.com\/related-docs\/prod_adapter_cards\/PB_Innova_Flex4_Lx_EN.pdf"},{"key":"e_1_3_1_87_2","unstructured":"Mellanox Technologies. 2019. Mellanox ASAP \\(^2\\) : Accelerated Switching and Packet Processing. Retrieved July 2026 from https:\/\/www.mellanox.com\/files\/doc-2020\/sb-asap2.pdf"},{"key":"e_1_3_1_88_2","first-page":"1","volume-title":"Proceedings of the HotOS\u201903.","author":"Mogul Jeffrey C.","year":"2003","unstructured":"Jeffrey C. Mogul. 2003. TCP Offload is a dumb idea whose time has come. In Proceedings of the HotOS\u201903.USENIX Association, Berkeley, CA, USA, 1 pages."},{"key":"e_1_3_1_89_2","unstructured":"napa:tech;. 2021. Link Programmable. Retrieved July 2026 from https:\/\/www.napatech.com\/products\/link-programmable\/"},{"key":"e_1_3_1_90_2","doi-asserted-by":"publisher","DOI":"10.1145\/3230543.3230560"},{"key":"e_1_3_1_91_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2021.3058217"},{"key":"e_1_3_1_92_2","unstructured":"NVIDIA Corporation. 2011. NVIDIA GPUDirect. Retrieved July 2026 from https:\/\/developer.nvidia.com\/gpudirect"},{"key":"e_1_3_1_93_2","unstructured":"NVIDIA Corporation. 2019. GPUDirect Storage: A Direct Path Between Storage and GPU Memory. Retrieved July 2026 from https:\/\/developer.nvidia.com\/blog\/gpudirect-storage\/"},{"key":"e_1_3_1_94_2","unstructured":"NVIDIA Corporation. 2020. Innova\u2122-2 Flex. Retrieved July 2026 from https:\/\/www.nvidia.com\/en-us\/networking\/ethernet\/innova-2-flex\/"},{"key":"e_1_3_1_95_2","unstructured":"NVIDIA Corporation. 2020. Networking Product Documentation \u2013 MLNX_OFED v23.10-0.5.5.0 \u2013 Scalable Functions. Retrieved July 2026 from https:\/\/networking-docs.nvidia.com\/mlnxofedswum\/23100550\/scalable-functions"},{"key":"e_1_3_1_96_2","unstructured":"NVIDIA Corporation. 2021. ConnectX\u00ae-6 Dx Ethernet SmartNIC. Retrieved July 2026 from https:\/\/www.mellanox.com\/files\/doc-2020\/pb-connectx-6-dx-en-card.pdf"},{"key":"e_1_3_1_97_2","unstructured":"NVIDIA Corporation. 2021. GPUDirect RDMA: Synchronization and Memory Ordering. Retrieved July 2026 from https:\/\/docs.nvidia.com\/cuda\/gpudirect-rdma\/#sync-behavior"},{"key":"e_1_3_1_98_2","unstructured":"NVIDIA Corporation. 2021. NVIDIA BlueField\u00ae-2 DPU. Retrieved July 2026 from https:\/\/www.nvidia.com\/content\/dam\/en-zz\/Solutions\/Data-Center\/documents\/datasheet-nvidia-bluefield-2-dpu.pdf"},{"key":"e_1_3_1_99_2","unstructured":"NVIDIA Corporation. 2021. NVLink and NVSwitch. Retrieved July 2026 from https:\/\/www.nvidia.com\/en-us\/data-center\/nvlink\/"},{"key":"e_1_3_1_100_2","unstructured":"NVIDIA Corporation. 2022. Howto Configure NVMe over Fabrics. Retrieved July 2026 from https:\/\/enterprise-support.nvidia.com\/s\/article\/howto-configure-nvme-over-fabrics"},{"key":"e_1_3_1_101_2","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPSW.2014.111"},{"key":"e_1_3_1_102_2","first-page":"515","volume-title":"2021 IFIP\/IEEE International Symposium on Integrated Network Management (IM)","author":"Pac\u00edfico Racyus D. G.","year":"2021","unstructured":"Racyus D. G. Pac\u00edfico, Matheus S. Castanho, Luiz F. M. Vieira, Marcos A. M. Vieira, Lucas F. S. Duarte, and Jos\u00e9 A. M. Nacif. 2021. Application Layer Packet Classifier in Hardware. In 2021 IFIP\/IEEE International Symposium on Integrated Network Management (IM). 515\u2013522."},{"key":"e_1_3_1_103_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.jalgor.2003.12.002"},{"key":"e_1_3_1_104_2","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2016.2548658"},{"key":"e_1_3_1_105_2","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00075"},{"key":"e_1_3_1_106_2","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446732"},{"key":"e_1_3_1_107_2","first-page":"949","volume-title":"Proceedings of the OSDI\u201923","author":"Pismenny Boris","year":"2023","unstructured":"Boris Pismenny, Adam Morrison, and Dan Tsafrir. 2023. ShRing: Networking with shared receive rings. In Proceedings of the OSDI\u201923. USENIX Association, Boston, MA, 949\u2013968."},{"key":"e_1_3_1_108_2","volume-title":"Proceedings of the NSDI 19","author":"Pontarelli Salvatore","year":"2019","unstructured":"Salvatore Pontarelli, Roberto Bifulco, Marco Bonola, Carmelo Cascone, Marco Spaziani, Valerio Bruschi, Davide Sanvito, Giuseppe Siracusano, Antonio Capone, Michio Honda, et\u00a0al. 2019. FlowBlaze: Stateful packet processing in hardware. In Proceedings of the NSDI 19. USENIX Association, Boston, MA."},{"key":"e_1_3_1_109_2","doi-asserted-by":"publisher","DOI":"10.1145\/2503210.2503288"},{"key":"e_1_3_1_110_2","doi-asserted-by":"publisher","DOI":"10.1109\/HiPC.2017.00037"},{"key":"e_1_3_1_111_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICPP.2013.17"},{"key":"e_1_3_1_112_2","doi-asserted-by":"publisher","DOI":"10.1145\/2996868"},{"key":"e_1_3_1_113_2","doi-asserted-by":"publisher","DOI":"10.1109\/ICSPCom.2015.7150618"},{"key":"e_1_3_1_114_2","doi-asserted-by":"publisher","DOI":"10.1145\/3445814.3446723"},{"key":"e_1_3_1_115_2","unstructured":"Yong Ren. 2018. High performance Cloud with Hardware Acceleration. Retrieved July 2026 from https:\/\/conferences.sigcomm.org\/events\/apnet2018\/slides\/yong.pdf"},{"key":"e_1_3_1_116_2","volume-title":"Proceedings of the USENIX ATC 12","author":"Rinta-aho Teemu","year":"2012","unstructured":"Teemu Rinta-aho, Mika Karlstedt, and Madhav P. Desai. 2012. The Click2NetFPGA Toolchain. In Proceedings of the USENIX ATC 12. USENIX, Boston, MA."},{"key":"e_1_3_1_117_2","doi-asserted-by":"publisher","DOI":"10.1145\/3289602.3293903"},{"key":"e_1_3_1_118_2","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2019.00053"},{"key":"e_1_3_1_119_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2017.40"},{"key":"e_1_3_1_120_2","first-page":"1005","volume-title":"Proceedings of the OSDI\u201923","author":"Sadok Hugo","year":"2023","unstructured":"Hugo Sadok, Nirav Atre, Zhipeng Zhao, Daniel S. Berger, James C. Hoe, Aurojit Panda, Justine Sherry, and Ren Wang. 2023. Ens\u014d: A streaming interface for NIC-application communication. In Proceedings of the OSDI\u201923. USENIX Association, Boston, MA, 1005\u20131025."},{"key":"e_1_3_1_121_2","doi-asserted-by":"publisher","DOI":"10.1145\/3617232.3624868"},{"key":"e_1_3_1_122_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2020.3016891"},{"key":"e_1_3_1_123_2","first-page":"69","volume-title":"Proceedings of the 13th USENIX Conference on Operating Systems Design and Implementation.","author":"Shan Yizhou","year":"2018","unstructured":"Yizhou Shan, Yutong Huang, Yilun Chen, and Yiying Zhang. 2018. LegoOS: A disseminated, distributed OS for hardware resource disaggregation. In Proceedings of the 13th USENIX Conference on Operating Systems Design and Implementation.USENIX Association, USA, 69\u201387."},{"key":"e_1_3_1_124_2","doi-asserted-by":"publisher","DOI":"10.17487\/RFC7252"},{"key":"e_1_3_1_125_2","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2015.12"},{"key":"e_1_3_1_126_2","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2016.7577319"},{"key":"e_1_3_1_127_2","doi-asserted-by":"publisher","DOI":"10.1145\/3342195.3387519"},{"key":"e_1_3_1_128_2","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3405897"},{"key":"e_1_3_1_129_2","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378509"},{"key":"e_1_3_1_130_2","unstructured":"ETSI\/SAGE Specification. 2011. Specification of the 3GPP Confidentiality and Integrity Algorithms 128-EEA3 and 128-EIA3. Document 2: ZUC Specification. Retrieved from https:\/\/www.gsma.com\/aboutus\/wp-content\/uploads\/2014\/12\/eea3eia3zucv16.pdf"},{"key":"e_1_3_1_131_2","doi-asserted-by":"publisher","DOI":"10.1147\/JRD.2018.2856978"},{"key":"e_1_3_1_132_2","volume-title":"Proceedings of the USENIX ATC 17","author":"Sultana Nik","year":"2017","unstructured":"Nik Sultana, Salvator Galea, David Greaves, Marcin Wojcik, Jonny Shipton, Richard Clegg, Luo Mai, Pietro Bressana, Robert Soul\u00e9, Richard Mortier, et\u00a0al. 2017. Emu: Rapid prototyping of networking services. In Proceedings of the USENIX ATC 17. USENIX Association, Santa Clara, CA."},{"key":"e_1_3_1_133_2","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2006.1639336"},{"key":"e_1_3_1_134_2","doi-asserted-by":"publisher","DOI":"10.1109\/SAMOS.2017.8344644"},{"key":"e_1_3_1_135_2","unstructured":"The kernel development community. 2019. Linux Networking Documentation. Retrieved July 2026 from https:\/\/www.kernel.org\/doc\/html\/latest\/networking\/index.html"},{"key":"e_1_3_1_136_2","unstructured":"The Linux Foundation Projects. 2018. DPDK Tools User Guides \u2013 dpdk-test-crypto-perf Application. Retrieved July 2026 from https:\/\/doc.dpdk.org\/guides\/tools\/cryptoperf.html"},{"key":"e_1_3_1_137_2","doi-asserted-by":"publisher","DOI":"10.1109\/CSCN.2017.8088623"},{"key":"e_1_3_1_138_2","doi-asserted-by":"publisher","DOI":"10.1109\/RECONFIG.2018.8641696"},{"key":"e_1_3_1_139_2","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378528"},{"key":"e_1_3_1_140_2","unstructured":"TRex Team. 2016. TRex: Realistic Traffic Generator. Retrieved July 2026 from https:\/\/trex-tgn.cisco.com\/"},{"key":"e_1_3_1_141_2","unstructured":"Animesh Trivedi. 2011. Remote Direct Memory Access (RDMA) 101 \u2013 Quick History Lesson and Introduction. Retrieved July 2026 from http:\/\/0x8086.blogspot.com\/2011\/11\/remote-direct-memory-access-rdma-101.html"},{"key":"e_1_3_1_142_2","first-page":"565","volume-title":"Proceedings of the NSDI 19","author":"\u00d0uki\u0107 Vojislav","year":"2019","unstructured":"Vojislav \u00d0uki\u0107, Sangeetha Abdu Jyothi, Bojan Karlas, Muhsen Owaida, Ce Zhang, and Ankit Singla. 2019. Is advance knowledge of flow sizes a plausible assumption?. In Proceedings of the NSDI 19. USENIX Association, Boston, MA, 565\u2013580."},{"key":"e_1_3_1_143_2","doi-asserted-by":"publisher","DOI":"10.1016\/j.micpro.2019.102861"},{"key":"e_1_3_1_144_2","doi-asserted-by":"publisher","DOI":"10.1109\/HOTI.2019.00017"},{"key":"e_1_3_1_145_2","doi-asserted-by":"publisher","DOI":"10.1145\/3651890.3672271"},{"key":"e_1_3_1_146_2","doi-asserted-by":"publisher","DOI":"10.1145\/3050220.3050234"},{"key":"e_1_3_1_147_2","volume-title":"Proceedings of the NSDI 22","author":"Wang Tao","year":"2022","unstructured":"Tao Wang, Xiangrui Yang, Gianni Antichi, Anirudh Sivaraman, and Aurojit Panda. 2022. Isolation mechanisms for high-speed packet-processing pipelines. In Proceedings of the NSDI 22. USENIX Association, Renton, WA."},{"key":"e_1_3_1_148_2","doi-asserted-by":"publisher","DOI":"10.1145\/3634916"},{"key":"e_1_3_1_149_2","doi-asserted-by":"publisher","DOI":"10.1109\/FPT.2016.7929186"},{"key":"e_1_3_1_150_2","doi-asserted-by":"publisher","DOI":"10.1145\/3503222.3507731"},{"key":"e_1_3_1_151_2","unstructured":"NVM Express Workgroup. 2019. NVM Express\u2122 over Fabrics Revision 1.1. Retrieved from https:\/\/nvmexpress.org\/wp-content\/uploads\/NVMe-over-Fabrics-1.1-2019.10.22-Ratified.pdf"},{"key":"e_1_3_1_152_2","unstructured":"Xilinx. 2021. Alveo U25 SmartNIC Accelerator Card. Retrieved July 2026 from https:\/\/www.xilinx.com\/products\/boards-and-kits\/alveo\/u25.html"},{"key":"e_1_3_1_153_2","unstructured":"Xilinx. 2021. UltraScale+ Devices Integrated 100G Ethernet Subsystem v3.1. Retrieved July 2026 from https:\/\/docs.amd.com\/r\/en-US\/pg203-cmac-usplus\/Introduction"},{"key":"e_1_3_1_154_2","unstructured":"Xilinx Inc.2017. SDNet PX Programming Language. Retrieved July 2026 from https:\/\/docs.amd.com\/v\/u\/en-US\/ug1016-px-programming"},{"key":"e_1_3_1_155_2","doi-asserted-by":"publisher","DOI":"10.1145\/3297858.3304024"},{"key":"e_1_3_1_156_2","first-page":"1","volume-title":"Proceedings of the 2nd USENIX Conference on Hot Topics in Cloud Computing.","author":"Zaharia Matei","year":"2010","unstructured":"Matei Zaharia, Mosharaf Chowdhury, Michael J. Franklin, Scott Shenker, and Ion Stoica. 2010. Spark: Cluster computing with working sets. In Proceedings of the 2nd USENIX Conference on Hot Topics in Cloud Computing.USENIX Association, USA, 1 pages."},{"key":"e_1_3_1_157_2","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2014.61"}],"container-title":["ACM Transactions on Computer Systems"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3822399","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,13]],"date-time":"2026-08-13T16:59:44Z","timestamp":1786640384000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3822399"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8,13]]},"references-count":156,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,8,31]]}},"alternative-id":["10.1145\/3822399"],"URL":"https:\/\/doi.org\/10.1145\/3822399","relation":{},"ISSN":["0734-2071","1557-7333"],"issn-type":[{"value":"0734-2071","type":"print"},{"value":"1557-7333","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8,13]]},"assertion":[{"value":"2024-07-18","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-05-29","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2026-08-13","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}