The Future of FPGA Acceleration in Datacenters and the Cloud
The DOI matches its Crossref record; 139 of the 143 references it cites were found in a registry or an index.
Abstract
In this article, we survey existing academic and commercial efforts to provide Field-Programmable Gate Array (FPGA) acceleration in datacenters and the cloud. The goal is a critical review of existing systems and a discussion of their evolution from single workstations with PCI-attached FPGAs in the early days of reconfigurable computing to the integration of FPGA farms in large-scale computing infrastructures. From the lessons learned, we discuss the future of FPGAs in datacenters and the cloud and assess the challenges likely to be encountered along the way. The article explores current architectures and discusses scalability and abstractions supported by operating systems, middleware, and virtualization. Hardware and software security becomes critical when infrastructure is shared among tenants with disparate backgrounds. We review the vulnerabilities of current systems and possible attack scenarios and discuss mitigation strategies, some of which impact FPGA architecture and technology. The viability of these architectures for popular applications is reviewed, with a particular focus on deep learning and scientific computing. This work draws from workshop discussions, panel sessions including the participation of experts in the reconfigurable computing field, and private discussions among these experts. These interactions have harmonized the terminology, taxonomy, and the important topics covered in this manuscript.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Citations
Cites
Spectre Attacks: Exploiting Speculative Execution
Kocher, P., Horn, J., Fogh, A., Genkin, D., Gruss, D., Haas, W., Hamburg, M., Lipp, M., Mangard, S., Prescher, T., et al. (2019) · 2019 IEEE Symposium on Security and Privacy (SP) · doi:10.1109/sp.2019.00002 · Cited by 1,780
2014 IEEE Symposium on Security and Privacy
(2014) · doi:10.1109/sp16517.2014 · Cited by 1,362
The 26th USENIX Security Symposium 2017
Αλεξόπουλος, Ν., Kiayias, A., Zacharias, T., & Talviste, R. (2017) · USENIX Security Symposium · Cited by 1,168
Containers and Cloud: From LXC to Docker to Kubernetes
Bernstein, D. (2014) · IEEE Cloud Computing · doi:10.1109/mcc.2014.51 · Cited by 1,098
Flipping bits in memory without accessing them: an experimental study of DRAM disturbance errors
Kim, Y., Daly, R., Kim, J., Fallin, C., Lee, J. H., Lee, D., Wilkerson, C., Lai, K., & Mutlu, O. (2014) · ACM SIGARCH Computer Architecture News · doi:10.1145/2678373.2665726 · Cited by 921
Towards predictable datacenter networks
Ballani, H., Costa, P., Karagiannis, T., & Rowstron, A. (2011) · Proceedings of the ACM SIGCOMM 2011 conference · doi:10.1145/2018436.2018465 · Cited by 750
A reconfigurable fabric for accelerating large-scale datacenter services
Putnam, A., Caulfield, A. M., Chung, E. S., Chiou, D., Constantinides, K., Demme, J., Esmaeilzadeh, H., Fowers, J., Gopal, G. P., Gray, J., et al. (2014) · ACM SIGARCH Computer Architecture News · doi:10.1145/2678373.2665678 · Cited by 693
OpenStack: Toward an Open-source Solution for Cloud Computing
Sefraoui, O., Aissaoui, M., & Eleuldj, M. (2012) · International Journal of Computer Applications · doi:10.5120/8738-2991 · Cited by 553
A Configurable Cloud-Scale DNN Processor for Real-Time AI
Fowers, J., Ovtcharov, K., Papamichael, M., Massengill, T., Liu, M., Lo, D., Alkalay, S., Haselman, M., Adams, L., Ghandi, M., et al. (2018) · 2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA) · doi:10.1109/isca.2018.00012 · Cited by 520
FINN- R: An End-to-End Deep-Learning Framework for Fast Exploration of Quantized Neural Networks
Blott, M., Preußer, T. B., Fraser, N. J., Gambardella, G., O’brien, K., Umuroglu, Y., Leeser, M., & Vissers, K. (2018) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3242897 · Cited by 446
Self-citation
Research works printed but not linked: 35
- 10.5555/874060.875374 · doi:10.5555/874060.875374
- ARM. 2010. AMBA 4 AXI4-Stream Protocol Specification. Technical Report. ARM.
- ARM. 2011. AMBA AXI and ACE Protocol Specification. Technical Report. ARM.
- Michaela Blott, Kimon Karras, Ling Liu, Kees Vissers, Jeremia Bär, and Zsolt István. 2013. Achieving 10Gbps line-rate key-value stores with FPGAs. In 5th USENIX Workshop on Hot Topics in Cloud Computing (HotCloud’13). USENIX Association, San Jose, CA. Retrieved from https://www.usenix.org/conference/hotcloud13/workshop-program/presentations/blott.
- 10.5555/2813749.2813757 · doi:10.5555/2813749.2813757
- 10.5555/1564913 · doi:10.5555/1564913
- 10.5555/2650280.2650362 · doi:10.5555/2650280.2650362
- 10.5555/3195638.3195647 · doi:10.5555/3195638.3195647
- 10.5555/3027041.3027052 · doi:10.5555/3027041.3027052
- Chris Conger, Ian Troxel, D. Espinoza, Vikas Aggarwal, and A. George. 2005. NARC: Network attached reconfigurable computing for high performance, network based applications. In 8th Annual International Conference on Military and Aerospace Programmable Logic Devices (MAPLD’05).
- Maik Ender, Amir Moradi, and Christof Paar. 2020. The unpatchable silicon: A full break of the bitstream encryption of Xilinx 7-series FPGAs. In 29th USENIX Security Symposium.
- 10.5555/3358807.3358838 · doi:10.5555/3358807.3358838
- Dieter Ernst. 2020. Competing in Artificial Intelligence Chips: China’s Challenge Amid Technology War. Centre for International Governance Innovation, Special Report.
- 10.5555/3307441.3307446 · doi:10.5555/3307441.3307446
- William Gayde. 2020. How Arm Came to Dominate the Mobile Market and It’s Coming for More, Much More. Retrieved from https://www.techspot.com/article/1989-arm-inside.
- Nicole Hemsothd. 2017. First in-depth look at Google’s new second-generation TPU. Retrieved from https://www.nextplatform.com/2017/05/17/first-depth-look-googles-new-second-generation-tpu/.
- 10.5555/2930611.2930639 · doi:10.5555/2930611.2930639
- 10.5555/3488766.3488822 · doi:10.5555/3488766.3488822
- 10.5555/3277203.3277276 · doi:10.5555/3277203.3277276
- Peter Loscocco and Stephen Smalley. 2001. Meeting critical security objectives with security-enhanced Linux. In Ottawa Linux Symposium. 115–134.
- Layong Larry Luo and T. E. G. Tencent. 2018. In Towards Converged SmartNIC Architecture for Bare Metal and Public Clouds at Tencent Scale 2nd Asia-Pacific Workshop on Networking (APNet 2018) August 2-3 2018 Beijing China .
- 10.5555/2971808.2972016 · doi:10.5555/2971808.2972016
- Christian Plessl. 2018. Bringing FPGAs to HPC production systems and codes. In H2RC’18 Workshop at Supercomputing (SC’18). DOI:https://doi.org/10.13140/RG.2.2.34327.42407 · doi:10.13140/rg.2.2.34327.42407
- 10.5555/647325.721663 · doi:10.5555/647325.721663
- Grigory Sapunov. 2021. Will ASIC Chips Become the Next Big Thing in AI? Retrieved from https://moorinsightsstrategy.com/will-asic-chips-become-the-next-big-thing-in-ai/.
- 10.5555/3195638.3195659 · doi:10.5555/3195638.3195659
- 10.5555/1204009 · doi:10.5555/1204009
- 10.5555/2978159 · doi:10.5555/2978159
- 10.5555/3437539.3437757 · doi:10.5555/3437539.3437757
- Naif Tarafdar and Paul Chow. 2019. libGalapagos: A software environment for prototyping and creating heterogeneous FPGA and CPU applications. In 6th International Workshop on FPGAs for Software Programmers (FSP’19).
- Malte Vesper, Dirk Kocha, and Khoa Phama. 2017. PCIeHLS: An OpenCL HLS framework. In 4th International Workshop on FPGAs for Software Programmers. VDE, 1–6.
- Greg Watson, Nick McKeown, and Martin Casado. 2006. NetFPGA: A tool for network research and education. In 2nd Workshop on Architectural Research Using FPGA Platforms (WARFP), Vol. 3.
- Xilinx Case Study. [n.d.]. Xilinx Powers Alibaba Cloud FaaS with AI Acceleration Solution for E-Commerce Business. Retrieved from https://www.xilinx.com/publications/powered-by-xilinx/xilinx-alibaba-case-study.pdf.
- Xilinx Corporation 2021. Virtex UltraScale+ FPGA Data Sheet: DC and AC Switching Characteristics.
- Wang Xu. 2018. Hardware acceleration over NFV in China Mobile. OPNFV Plugfest.
Cited by
Edge AI for Internet of Energy: Challenges and perspectives
Himeur, Y., Sayed, A. N., Alsalemi, A., Bensaali, F., & Amira, A. (2023) · Internet of Things · doi:10.1016/j.iot.2023.101035 · Cited by 130
Empowering Cloud Computing With Network Acceleration: A Survey
Rosa, L., Foschini, L., & Corradi, A. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3377531 · Cited by 52
A Survey on FPGA-Based Heterogeneous Clusters Architectures
Florian, W., Crespo, M. L., Cicuttin, A., & Carrato, S. (2023) · IEEE Access · doi:10.1109/access.2023.3288431 · Cited by 25
“An in-depth study of FPGA cloud computing architectures is available in [79] and [80].”
BackgroundGuo, A., Hao, Y., Wu, C., Haghi, P., Pan, Z., Si, M., Tao, D., Li, A., Herbordt, M., & Geng, T. (2023) · International Conference on Supercomputing · doi:10.1145/3577193.3593724 · Cited by 37
Self-citationA Survey of FPGA Optimization Methods for Data Center Energy Efficiency
Tibaldi, M., & Pilato, C. (2023) · IEEE Transactions on Sustainable Computing · doi:10.1109/tsusc.2023.3273852 · Cited by 22
“Such a deployment needs fewer FPGAs; this typically translates to easier management, the lower total cost of ownership (TCO), and smaller average node sizes [112].”
BackgroundInfluentialDistributed large-scale graph processing on FPGAs
Sahebi, A., Barbone, M., Procaccini, M., et al. (2023) · Journal Of Big Data · doi:10.1186/s40537-023-00756-x · Cited by 15
“The use of FPGAs in combination with Hadoop for large graph processing is an emerging field that has gained attention in recent years [11, 53].”
MethodRecent advances on reliability of FPGAs in a radiation environment
Liu, Z., Lu, Z., Huang, L., et al. (2024) · Microelectronics Journal · doi:10.1016/j.mejo.2024.106176 · Cited by 15
Multi-Tenant Cloud FPGA: A Survey on Security, Trust, and Privacy
Ahmed, M. K., Panoff, M., Mbongue, J. M., Saha, S. K., Tchinda, E. N., Mbua, P., & Bobda, C. (2025) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3713078 · Cited by 34
Self-citationTowards Secure Runtime Customizable Trusted Execution Environment on FPGA-SoC
Wang, Y.-L., Chang, X., Zhu, H., et al. (2024) · IEEE Transactions on Computers · doi:10.1109/tc.2024.3355772 · Cited by 14
“Cloud Service Providers (CSP), such as Microsoft, Alibaba, Amazon and Huawei, are adopting FPGA with unmatched capabilities and intrinsic advantages into their infrastructures [5].”
BackgroundHMT: A Hardware-centric Hybrid Bonsai Merkle Tree Algorithm for High-performance Authentication
Shadab, R. M., Zou, Y., Gandham, S., et al. (2023) · ACM Transactions on Embedded Computing Systems · doi:10.1145/3595179 · Cited by 12
“The rapid emergence of FPGA devices for security and acceleration in blockchain and cloud applications such as in Amazon AWS [5, 23, 31] and state-of-the-art FPGA-driven secure embedded platforms such as ARES [38] have made it crucial to have a low-latency, high throughput integrity authentication mechanism directly i…”
BackgroundFASDA: An FPGA-Aided, Scalable and Distributed Accelerator for Range-Limited Molecular Dynamics
Wu, C., Geng, T., Guo, A., et al. (2023) · 2012 International Conference for High Performance Computing, Networking, Storage and Analysis · doi:10.1145/3581784.3607100 · Cited by 11
Self-citationFPGA-Based Sensors for Distributed Digital Manufacturing Systems: A State-of-the-Art Review
Khan, L., Isanaka, S. P., & Liou, F. (2024) · Sensors · doi:10.3390/s24237709 · Cited by 12
“Synchronizing several FPGA-based sensors in a scattered digital factory can also result in processing delays, bandwidth constraints, and data inconsistencies, contributing to lessening the system efficiency.”
Coyote v2: Raising the Level of Abstraction for Data Center FPGAs
Ramhorst, B., Korolija, D., Heer, M. J., Dann, J., Liu, L., & Alonso, G. (2025) · ACM SIGOPS Symposium on Operating Systems Principles (SOSP) · doi:10.1145/3731569.3764845 · Cited by 24
Enabling VirtIO Driver Support on FPGAs
Bandara, S., Sanaullah, A., Tahir, Z., et al. (2022) · doi:10.1109/h2rc56700.2022.00006 · Cited by 9
Self-citationRajarathnam, R. S., Jiang, Z., Iyer, M. A., et al. (2023) · International Symposium on Physical Design · doi:10.1145/3569052.3571881 · Cited by 8
“3571881 1 INTRODUCTION The evolution of modern field programmable gate arrays (FPGAs) in heterogeneity, capability, and capacity allows their usage in varied applications such as integrated circuit (IC) prototyping and as accelerators in data centers and the cloud [4, 13].”
BackgroundA Visionary Look at the Security of Reconfigurable Cloud Computing
Stojilović, M., Rasmussen, K., Regazzoni, F., Tahoori, M. B., & Tessier, R. (2023) · Proceedings of the IEEE · doi:10.1109/jproc.2023.3330729 · Cited by 9
“2 corresponds to the single-node accelerator model, deployed by a number of cloud providers [7], such as AWS, Huawei, Baidu, Tencent, Nimbix, and Alibaba.”
BackgroundMethodInfluentialSelf-citationInkollu, U. M. R., & Sastry, J. K. R. (2024) · Earth Science Informatics · doi:10.1007/s12145-024-01242-5 · Cited by 8
“While efficient in the early days of cloud computing, contemporary applications brought in unstable and dynamic workloads, particularly in high-throughput fields like earth science (Bobda, et al. 2022).”
MESSI: Task Mapping and Scheduling Strategy for FPGA-based Heterogeneous Real-Time Systems
Ahmadi-Pour, S., Saha, S., McDonald-Maier, K. D., & Drechsler, R. (2025) · ACM Transactions on Design Automation of Electronic Systems · doi:10.1145/3715323 · Cited by 8
“With the increasing complexity level of high-performance computing and RT embedded systems, current heterogeneous computing systems are employing FPGAs along with CPUs and GPUs to overcome existing limitations [3, 6].”
Development of Electronic-Photonic 3D System in Package: Architecture, Integration, and Scaling
Bhandari, J. K., Rajeswari, D. S., Ramana, P. V., et al. (2023) · 2022 IEEE 72nd Electronic Components and Technology Conference (ECTC) · doi:10.1109/ectc51909.2023.00327 · Cited by 7
Gookyi, D. A. N., Lee, E., Kim, K., Jang, S., & Lee, S.-S. (2023) · Sensors · doi:10.3390/s23052380 · Cited by 10
“shorter time-to-market of products [15].”
A Framework for Neural Network Inference on FPGA-Centric SmartNICs
Guo, A., Geng, T., Zhang, Y., et al. (2022) · International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl57034.2022.00071 · Cited by 7
“Dozens of commercial FPGA-based SmartNICs have been released, including from AMD and Intel, e.g., [8]–[10]; surveys include [11], [12].”
Self-citationServerless FPGA: Work-In-Progress
Maschi, F., Korolija, D., & Alonso, G. (2023) · Workshop on SErverless Systems, Applications and MEthodologies (SESAME) · doi:10.1145/3592533.3592804 · Cited by 11
RPUGuard: Real-Time Processing Unit Virtualization for Mixed-Criticality Applications
Cinque, M., Tommasi, G. D., Dubbioso, S., & Ottaviano, D. (2022) · European Dependable Computing Conference · doi:10.1109/edcc57035.2022.00025 · Cited by 14
“Virtualization of FPGAs is also a very complex area of research [28] [29].”
BackgroundChen, J., Yang, S., Wang, Y., et al. (2024) · Cluster Computing · doi:10.1007/s10586-024-04401-x · Cited by 6
“FPGA [22] ( Field-Programmable Gate Array) is an integrated circuit that can be reconfigured at the hardware level.”
Automated parallel execution of distributed task graphs with FPGA clusters
Ruiz, J. M. D. H., Álvarez, C., Jiménez-González, D., et al. (2024) · Future Generation Computer Systems · doi:10.1016/j.future.2024.06.041 · Cited by 6
Self-citationChang, Y., & Guo, Z. (2024) · IEEE Transactions on Very Large Scale Integration (VLSI) Systems · doi:10.1109/tvlsi.2024.3436017 · Cited by 6
Single Event Effects Assessment of UltraScale+ MPSoC Systems Under Atmospheric Radiation
Agiakatsikas, D., Foutris, N., Sari, A., Vlagkoulis, V., Souvatzoglou, I., Psarakis, M., Ye, R., Goodacre, J., Luján, M., Kastriotou, M., et al. (2023) · IEEE Transactions on Reliability · doi:10.1109/tr.2023.3312548 · Cited by 11
“Given that the demand for FPGAs in cloud and data centre facilities will increase in the upcoming decade, and the likelihood of NSEU-related failures may become a significant problem [17].”
BackgroundTaming the Elephants: Affordable Flow Length Prediction in the Data Plane
Azorin, R., Monterubbiano, A., Castellano, G., Gallo, M., Pontarelli, S., & Rossi, D. (2024) · Proceedings of the ACM on Networking · doi:10.1145/3649473 · Cited by 6
“However, we still include results on UNI, also used in [51], as it enables us to estimate the applicability of our approach on datacenter-like traffic.”
Algorithms for Fast Spiking Neural Network Simulation on FPGAs
Lindqvist, B. A., & Podobas, A. (2024) · IEEE Access · doi:10.1109/access.2024.3479933 · Cited by 5
“Potential drawbacks aside, researchers have deployed FPGAs for performance sensitive workloads, with promising results [36], [37].”
BackgroundTurboHE: Accelerating Fully Homomorphic Encryption Using FPGA Clusters
Liao, H., Elmohr, M. A., Dong, X., Qian, Y., Yang, W., Shang, Z., & Tan, Y. (2023) · IEEE International Parallel and Distributed Processing Symposium (IPDPS) · doi:10.1109/ipdps54959.2023.00084 · Cited by 6
“Note that there already exists FaaS in market that provides multiple FPGAs in the form of a cluster to accelerate other applications [18].”
BackgroundInstruction-Level Power Side-Channel Leakage Evaluation of Soft-Core CPUs on Shared FPGAs
Glamočanin, O., Shrivastava, S., Yao, J., Ardo, N., Payer, M., & Stojilović, M. (2023) · Journal of Hardware and Systems Security · doi:10.1007/s41635-023-00135-1 · Cited by 12
“The challenges are numerous, such as partitioning the FPGA resources among multiple users, providing communication protocols between the host virtual machine and the accelerators, and ensuring proper physical and logical isolation between the tenants [15].”
BackgroundMahmoud, D. G., Shokry, B., Lenders, V., Hu, W., & Stojilović, M. (2024) · IEEE Access · doi:10.1109/access.2024.3353134 · Cited by 8
“Through the shell management, the CSP controls resource allocation and user isolation [7].”
BackgroundLow-Power Design of Advanced Image Processing Algorithms under FPGA in Real-Time Applications
Wang, Y. (2024) · IEEE International Conference on Power, Electronics and Computer Applications (ICPECA) · doi:10.1109/icpeca60615.2024.10471036 · Cited by 6
“Researchers have used FPGA as an implementation platform for neural networks and achieved good results, especially in terms of speed, power consumption, etc. [9-10].”
MethodHigh-Precision Time Synchronization Based on Timestamp Mapping in Datacenter Networks
Li, L., Chen, B., Duan, D., et al. (2025) · Electronics · doi:10.3390/electronics14030610 · Cited by 5
AI Accelerators for Cloud and Server Applications
Shrestha, R., Bajracharya, R., Mishra, A., et al. (2023) · Springer eBooks · doi:10.1007/978-3-031-22170-5_3 · Cited by 5
Cirici, S., Paolino, M., & Raho, D. (2023) · International Conference on Computer, Information and Telecommunication Systems · doi:10.1109/cits58301.2023.10188786 · Cited by 5
“Academic studies on the virtualization of FPGA resources are abundant in the research community [7], [8], covering a wide range of FPGA types (including PCIe, network, and SoC) and methodologies.”
BackgroundMachine Learning Across Network-Connected FPGAs
Diaconu, D., Xie, Y., Gungor, M., Handagala, S., Lin, X., & Leeser, M. E. (2023) · IEEE High Performance Extreme Computing Conference · doi:10.1109/hpec58863.2023.10363454 · Cited by 6
“The main purpose of FPGAs in the cloud is to accelerate different time-consuming tasks as they can achieve better performance for lower power consumption compared to CPUs and GPUs [23].”
BackgroundSelf-citationComputing Acceleration to Genome-Wide Association Study Based on CPU/FPGA Heterogeneous System
Liu, X., Wang, R., Shi, C., Zou, C., & Zhu, W. (2023) · ACM SIGAPP Applied Computing Review · doi:10.1145/3642964.3642966 · Cited by 5
“Different from GPU, FPGA has low power and programmability, and can also provide high computing parallelism [2].”
BackgroundBartzoudis, N., Fernández, J. R., López‐Bueno, D., et al. (2024) · Electronics · doi:10.3390/electronics13040701 · Cited by 5
High-efficiency Compressor Trees for Latest AMD FPGAs
Hoßfeld, K. J., Damsgaard, H. J., Nurmi, J., Blott, M., & Preußer, T. B. (2024) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3645097 · Cited by 6
“This trend is expected to continue as application diversity and throughput demands grow [4].”
BackgroundDoS-FPGA: Denial of Service on Cloud FPGAs via Coordinated Power Hammering
Nassar, H., Machauer, P., Bauer, L., Gnad, D. R. E., Tahoori, M. B., & Henkel, J. (2024) · doi:10.1145/3676536.3676843 · Cited by 6
“However, in contrast to multi-tenancy that is already deployed for classical computing devices or GPUs, multi-tenancy for FPGAs faces several challenges [15, 16].”
FCsN: A FPGA-Centric SmartNIC Framework for Neural Networks
Guo, A., Geng, T., Zhang, Y., et al. (2022) · IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm53951.2022.9786193 · Cited by 5
Self-citationA First Step towards Support for MPI Partitioned Communication on SYCL-programmed FPGAs
Christgau, S., Knaust, M., & Steinke, T. (2022) · doi:10.1109/h2rc56700.2022.00007 · Cited by 5
“This matches well with disaggregated resources where the FPGA can directly communicate with the network [12], [13].”
BackgroundMirhosseini, M., Fazlali, M., Fallah, M. K., & Lee, J. (2023) · The Journal of Supercomputing · doi:10.1007/s11227-023-05109-2 · Cited by 4
“…of digital circuits for manufacturing efficient hardware has various applications such as Internet of things [12, 13], cloud/fog/edge computing [2, 14], embedded systems [15–17], optical flow [18, 19], machine learning [20–22], bio-informatics [23, 24], streaming [25], scientific computing [26],…”
BackgroundDivi, A., & Binzaid, S. (2025) · Solar · doi:10.3390/solar5020024 · Cited by 4
“Specifically, if the outputs corresponding to the OC and SC states (B and C signals) are applied to the XOR gate, it will produce a HIGH output when there is a discrepancy between the two signals [38–40].”
Guo, A., Hao, Y., Yao, X., Yang, S., Huang, J., Geng, T., & Herbordt, M. (2025) · International Conference on Supercomputing · doi:10.1145/3721145.3729514 · Cited by 6
Self-citationPerformance Evaluation of VirtIO Device Drivers for Host-FPGA PCIe Communication
Bandara, S., Sanaullah, A., Tahir, Z., et al. (2024) · IEEE International Parallel and Distributed Processing Symposium Workshops (IPDPSW) · doi:10.1109/ipdpsw63119.2024.00043 · Cited by 4
Self-citationFPGA innovation research in the Netherlands: present landscape and future outlook
Alachiotis, N., van den Belt, S., van der Vlugt, S., van der Walle, R., Safari, M., Forlin, B. E., Matteis, T. D., Al-Ars, Z., Jordans, R., de Almeida, A. J. S., et al. (2025) · Frontiers in High Performance Computing · doi:10.3389/fhpcp.2025.1572844 · Cited by 3
SAPER-AI accelerator: a systolic array-based power-efficient reconfigurable AI accelerator
Muslim, F. B., Inayat, K., Siddiqi, M. Z., Khan, S., Mahmood, T., & Islam, I. U. (2025) · Frontiers of Information Technology & Electronic Engineering · doi:10.1631/fitee.2400867 · Cited by 3
“Several such accelerators have been proposed on both sides of the DL continuum, i.e., the cloud (Bobda et al., 2022; Li et al., 2023) and the edge (Seshadri et al., 2022; Loh et al., 2025).”
µ-VF: Enabling Virtualization of Embedded FPGAs
Bucaria, V. A., Longo, F., Merlino, G., & Restuccia, F. (2025) · Proceedings of the ACM on Measurement and Analysis of Computing Systems · doi:10.1145/3771581 · Cited by 3
Optimizing AI Performance in Industry: A Hybrid Computing Architecture Approach Based on Big Data
Dewi, M. U., Santoso, L., & Santoso, A. B. (2024) · Journal of Technology Informatics and Engineering · doi:10.51903/jtie.v3i3.201 · Cited by 5
“Additionally, specialized hardware like FPGAs can accelerate specific tasks that require intensive processing (Bobda et al., 2022) .”
Harmonia: A Unified Framework for Heterogeneous FPGA Acceleration in the Cloud
Li, L., Pan, H., Wan, X., et al. (2025) · ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3676641.3716259 · Cited by 3
Hardware-Accelerated Edge AI Orchestration on the Multi-Tier Edge-to-Cloud Continuum
Palomares, J., Coronado, E., Tzenetopoulos, A., et al. (2025) · Journal of Network and Systems Management · doi:10.1007/s10922-025-09959-4 · Cited by 3
“Broader surveys [56–58] report throughput improvements of up to 20 × when AI models are executed on specialized accelerators, including FPGAs and TPUs.”
Memory Isolation for Multi-Tenant Data Integrity in Cloud MPSoC FPGAs
Donchez, S., & Wang, X. (2022) · IEEE Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON) · doi:10.1109/iemcon56893.2022.9946490 · Cited by 3
“Accordingly, it should come as no surprise that, as much of the world pivots from on-site data centers and computing resources to hybrid or cloud based platforms, more and more Cloud Service Providers (CSPs) are incorporating FPGAs in their cloud data centers [2].”
BackgroundFlowPix: Accelerating Image Processing Pipelines on an FPGA Overlay using a Domain Specific Compiler
Choudhury, Z., Gulati, A., & Purini, S. (2023) · ACM Transactions on Architecture and Code Optimization · doi:10.1145/3629523 · Cited by 3
A Tenant Side Compilation Solution for Cloud FPGA Deployment
Panoff, M., Ahmed, M. K., Wang, H., et al. (2023) · International Conference on Field-Programmable Technology (FPT) · doi:10.1109/icfpt59805.2023.00007 · Cited by 3
Self-citationData-driven modeling of reconfigurable multi-accelerator systems under dynamic workloads
Encinas, J., Rodríguez, A., Otero, A., & de la Torre, E. (2024) · Microprocessors and Microsystems · doi:10.1016/j.micpro.2024.105050 · Cited by 6
Fuchs, M., Krause, H. A., Muscheid, T., et al. (2024) · IEEE Transactions on Nuclear Science · doi:10.1109/tns.2024.3446309 · Cited by 3
“…utilizing various computational engines, including FPGAs, central processing units (CPUs), graphics processing units (GPUs), and artificial intelligence (AI) specific processors, are frequently deployed in data centers of major providers, such as Alibaba, Amazon, Baidu, Huawei, and Microsoft [13].”
Swift: A Multi-FPGA Framework for Scaling Up Accelerated Graph Analytics
Jaiyeoba, O., Mughrabi, A. T., Baradaran, M., et al. (2024) · International Conference on Field-Programmable Technology (FPT) · doi:10.1109/icfpt64416.2024.11113456 · Cited by 3
“…PCIe-connected FPGAs (multi-FPGA frameworks) exhibit lower latency compared to network-connected FPGAs (single-FPGA frameworks) due to fast/high bandwidth on-chip HBM memory coupled with PCIe DMA, providing direct access to host memory without the need to navigate the network stack [42] [43] [44].”
NeuroVM: Dynamic Neuromorphic Hardware Virtualization
Işık, M., Naoukin, J., & Dikmen, I. C. (2024) · doi:10.1109/igsc64514.2024.00023 · Cited by 2
“However, with the increasing complexity of computational tasks, particularly in artificial intelligence and machine learning, new paradigms for computing are needed [1]–[5].”
A Covert and Efficient Attack on FPGA Cloud Based on Adaptive RON
Tang, S., Wang, J., Chen, Z., & Guo, S. (2025) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2025.3550568 · Cited by 2
“To defend against attacks, the cloud computing platform proposes protective measures in three phases that are design, compilation [19], [20] and runtime [21], [22].”
E2C: A Visual Simulator to Reinforce Education of Heterogeneous Computing Systems
Mokhtari, A., Rawls, D., Huynh, T., Green, J., & Salehi, M. A. (2023) · IEEE International Symposium on Parallel & Distributed Processing, Workshops and Phd Forum · doi:10.1109/ipdpsw59300.2023.00052 · Cited by 3
“Hyperscalers such as AWS and Microsoft Azure provide computing services ranging from general-purpose X86-based and ARM-based machines to FPGAs and accelerators [1].”
BackgroundInfluentialExploring FPGA Acceleration for Distributed Serverless Computing
Yang, Z., & Fahmy, S. A. (2023) · International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl60245.2023.00059 · Cited by 2
“FPGAs also have a power profile and form factor that can support a wide range of deployment scenarios when networked [4], [5].”
BackgroundA Covert Attack Method Against FPGA Clouds
Tang, S., Wang, J., Chen, Z., et al. (2023) · Proceedings - Asian Test Symposium/Proceedings · doi:10.1109/ats59501.2023.10318003 · Cited by 2
“FPGA cloud vendors protect FPGA clouds from three stages, i.e., RTL design, compilation and operation [20], [21].”
MethodPDNSig: Identifying Multi-Tenant Cloud FPGAs with Power Distribution Network-Based Signatures
Zhu, H., Cao, W., & Zhang, X. (2023) · doi:10.1109/iccad57390.2023.10323545 · Cited by 2
“Particularly, they have been widely used by modern data centers, namely cloud FPGAs , for cloud-based applications such as machine learning [1]–[3] and cryptography [4], [5].”
MethodDomain Isolation and Access Control in Multi-tenant Cloud FPGAs
Bobda, C., Mbongue, J. M., Saha, S. K., & Ahmed, M. K. (2023) · Springer eBooks · doi:10.1007/978-3-031-45395-3_2 · Cited by 2
Self-citationEnabling high-level parallel programming on multi-FPGA clusters
Ruiz, J. M. D. H., Álvarez, C., Jiménez-González, D., et al. (2024) · International Symposium on Highly Efficient Accelerators and Reconfigurable Technologies (HEART) · doi:10.1145/3665283.3665292 · Cited by 2
“There are also surveys and studies of current cluster and cloud FPGA architectures [4, 25, 27].”
BackgroundMemory Scraping Attack on Xilinx FPGAs: Private Data Extraction from Terminated Processes
Madabhushi, B., Kundu, S. K., & Holcomb, D. E. (2024) · doi:10.23919/date58400.2024.10546527 · Cited by 2
“When offloading compute intensive tasks from host processing unit (PU) to FPGA for acceleration, the FPGA’s local DRAM is used to temporarily store [15] and reuse data before returning results to the host.”
MethodBetter Together: Combining Analytical and Annealing Methods for FPGA Placement
Rajarathnam, R. S., Thurmer, K., Betz, V., et al. (2024) · International Conference on Field Programmable Logic and Applications (FPL) · doi:10.1109/fpl64840.2024.00016 · Cited by 2
Parallel Accurate Minifloat MACCs for Neural Network Inference on Versal FPGAs
Damsgaard, H. J., Hoßfeld, K. J., Nurmi, J., & Preußer, T. B. (2024) · IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · doi:10.1109/tcad.2024.3511343 · Cited by 10
“The dot product is one such operation whose efficiency is key to applications that rely on matrix multiplication, such as neural network inference, data mining, and cryptography [4].”
Application of Network Calculus Models to Heterogeneous Streaming Applications
Faber, C. J., & Chamberlain, R. D. (2025) · International Journal of Networking and Computing · doi:10.15803/ijnc.15.1_51 · Cited by 2
FPGA-Based Large-Scale Sorting with Optimized Bandwidth Utilization
Sun, M., Xie, G., Zhang, F., et al. (2025) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3716392 · Cited by 2
LVMScissor: Split and Schedule Large Vision Model Inference on Mobile Edges via Salp Swarm Algorithm
Liu, Y., Lu, R., & Wang, D. (2025) · IEEE Transactions on Mobile Computing · doi:10.1109/tmc.2025.3550519 · Cited by 6
FPGA-Accelerated VXLAN Chaining for Partially Reconfigurable VNFs in Heterogeneous Data Centers
Chang, Y., & Guo, Z. (2025) · IEICE Transactions on Communications · doi:10.23919/transcom.2024ebp3197 · Cited by 2
An Overview of AI Workload Optimization Techniques
Panchumarthy, R., & Benala, T. R. (2025) · Artificial intelligence-enhanced software and systems engineering · doi:10.1007/978-3-031-88188-6_12 · Cited by 7
FARMER: An Online-Learning Driven Methodology for Workload Consolidation on Large FPGAs
Montanaro, G., Trovò, F., & Zoni, D. (2025) · IEEE International Symposium on Circuits and Systems proceedings · doi:10.1109/iscas56072.2025.11043238 · Cited by 2
Fault Recovery from Multi-Tenant FPGA Voltage Attacks
Moini, S., Kansagara, D., Holcomb, D. E., et al. (2023) · ACM Great Lakes Symposium on VLSI (GLSVLSI) · doi:10.1145/3583781.3590246 · Cited by 2
“…detect safe … Additionally, although the ability of multiple independent customers to share a commercial cloud FPGA is not currently supported, it is expected in the near future [10]. s i+1 s i+2 s i s i+k1 Thus, run-time protection against multi-tenant voltage attacks is an important concern [3].”
BackgroundSelf-citationPentimento: Data Remanence in Cloud FPGAs
Drewes, C., Weng, O., Meza, A., et al. (2024) · ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS) · doi:10.1145/3620665.3640355 · Cited by 2
“Our attack is a single-tenant temporal side channel – state within the FPGA that is not wiped correctly, or it is impossible to remove between subsequent users [9].”
BackgroundDesign and Development of an FPGA-Based Real-Time Reconfigurable Computing Platform
Kidav, J. U., & Mohan, V. (2023) · Lecture notes in networks and systems · doi:10.1007/978-981-99-1699-3_24 · Cited by 1
Mahmoud, D. G., Glamočanin, O., Regazzoni, F., et al. (2023) · Springer eBooks · doi:10.1007/978-3-031-45395-3_5 · Cited by 1
Countermeasures Against Voltage Attacks in Multi-tenant FPGAs
Moini, S., Provelengios, G., Holcomb, D. E., et al. (2023) · Springer eBooks · doi:10.1007/978-3-031-45395-3_10 · Cited by 1
Self-citationYe, Q., Wang, Y., Wu, C., Zheng, M., Lin, L., & Zhou, F. (2024) · International Conference on Electronic Engineering and Information Systems (EEISS) · doi:10.1109/eeiss62553.2024.00011 · Cited by 1
Security Risks Due to Data Persistence in Cloud FPGA Platforms
Zhang, Z., Madabhushi, B., Kundu, S. K., et al. (2024) · Conference proceedings · doi:10.1109/mwscas60917.2024.10658894 · Cited by 1
“Cloud FPGAs generally are attached to substantial DRAM storage that is either directly attached to the devices or accessible via a system-level bus [2] (e.g., PCIe).”
BackgroundMethodSelf-citationAggregate Encryption Individual Decryption for FPGA Bitstream Protection on Cloud
Debnath, M., Guha, K., Saha, D., et al. (2024) · doi:10.1109/seed61283.2024.00025 · Cited by 1
“Major cloud providers have integrated FPGA reconfigurability into user applications [4].”
BackgroundMethodRealizing Network-Attached FPGAs in the Cloud
Leeser, M. E., Handagala, S., Diaconu, D., et al. (2024) · Computing in Science & Engineering · doi:10.1109/mcse.2024.3482068 · Cited by 1
Self-citationOptimizing Energy Consumed by Analytics in the Cloud
ORDÓÑEZ, C. R., & Macyna, W. (2024) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata62323.2024.10825559 · Cited by 1
“In the last few years, FPGAs have been used in the cloud for diverse applications (see [5]).”
Leveraging CPU-FPGA Co-design for Matrix Profile Computation
Huseynli, F., Raoofy, A., & Schulz, M. (2025) · Communications in computer and information science · doi:10.1007/978-3-031-80084-9_9 · Cited by 1
Dynamic Function Exchange in FPGA to Redefine RISC-V Multicore Architectures at Runtime
Alves, T. S., & Bonato, V. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-87995-1_14 · Cited by 1
AutoClock: Automated Clock Management for Power-Efficient HLS Designs on FPGAs
Liang, J., Du, L., Zhou, X., et al. (2025) · Proceedings - ACM IEEE Design Automation Conference · doi:10.1109/dac63849.2025.11133279 · Cited by 1
F3: An FPGA-accelerated FaaS Framework
Mainas, C., Lambeck, M., Scheufler, B., et al. (2025) · ACM International Symposium on High-Performance Parallel and Distributed Computing (HPDC) · doi:10.1145/3731545.3731582 · Cited by 1
Mao, X., Zhou, Q., Chen, X., et al. (2025) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3779437 · Cited by 1
“Field-programmable gate arrays (FPGAs) are known for energy efficiency and 3:2 X. Mao et al. reconfigurable parallel architectures, making them widely used in computationally intensive fields such as neural network acceleration [1], homomorphic encryption [2], and big data processing [3].”
Electromagnetic Side-Channel Attack on a SoC-FPGA Fingerprint Recognition System
Tejeda, S., Román, R., Arjona, R., et al. (2025) · Conference on Design of Circuits and Integrated Systems · doi:10.1109/dcis67520.2025.11281924 · Cited by 1
“FPGAs are adopted by Cloud Service Providers (CSP), such as Microsoft, Alibaba, Amazon and Huawei [11].”
vFPGA: Towards Sub-µs Reconfiguration via 3D FPGA and Packaging Co-Design
Cherukuri, N. K., Nag, S., Nalla, P. S., Kola, A. K., Gadireddi, C. S., Dai, K., Seo, J.-S., Fang, Z., Zhang, J. J., & Cao, Y. K. (2026) · ACM/SIGDA International Symposium on Field-Programmable Gate Arrays (FPGA) · doi:10.1145/3748173.3779206 · Cited by 3
“They have become a versatile computing platform across cloud infrastructure, embedded systems, networking, adaptive deep learning, and scientific computing [1, 5, 14, 29, 37, 43].”
Beneath the fabric: A survey on threats and opportunities in reconfigurable systems
Pereira, S., Gomes, T., Cabral, J., & Pinto, S. (2026) · Journal of Systems Architecture · doi:10.1016/j.sysarc.2026.103779 · Cited by 1
COPA Use Case: Distributed Secure Joint Computation
Patel, R., Haghi, P., Jain, S., et al. (2022) · IEEE International Symposium on Field-Programmable Custom Computing Machines (FCCM) · doi:10.1109/fccm53951.2022.9786156 · Cited by 1
Self-citationTask Mapping and Scheduling in FPGA-based Heterogeneous Real-time Systems: A RISC-V Case-Study
Ahmadi-Pour, S., Saha, S., Herdt, V., et al. (2022) · Euromicro Conference on Digital System Design (DSD) · doi:10.1109/dsd57027.2022.00027 · Cited by 1
“Withthe increasing complexity level of high-performance computing and real-time embedded systems, current heterogeneous computing systems are employing FPGAs along with CPUs and GPUs to overcome existing limitations [1], [8].”
BackgroundOrganization of FPGA-based Devices in Distributed Systems
Maidan, M., & Melnyk, A. (2023) · International Journal of Computing · doi:10.47839/ijc.22.3.3231 · Cited by 1
H-Storm: A Hybrid CPU-FPGA Architecture to Accelerate Apache Storm
Nasiri, H., Darjani, A., Kavand, N., & Goudarzi, M. (2023) · Journal of Grid Computing · doi:10.1007/s10723-023-09692-9 · Cited by 1
On the Malicious Potential of Xilinx’s Internal Configuration Access Port (ICAP)
Albartus, N., Ender, M., Möller, J.-N., Fyrbiak, M., Paar, C., & Tessier, R. (2023) · ACM Transactions on Reconfigurable Technology and Systems · doi:10.1145/3633204 · Cited by 8
“This category of analysis encompasses various types, including simulation-based and on-chip methods [68].”
BackgroundSelf-citationA Survey on Fault-Tolerance Methods for SRAM-Based FPGAs in Radiation Environments
Lu, Z., Zhao, Q. L., Chen, Q., et al. (2023) · Proceedings - Asian Test Symposium/Proceedings · doi:10.1109/ats59501.2023.10318028 · Cited by 1
“Over the past few decades, the Field Programmable Gate Array (FPGA) technology has rapidly grown in terms of flexibility and computing power [1].”
Background
Metadata
- Authors
- Bobda, C., Mbongue, J. M., Chow, P., Ewais, M., Tarafdar, N., Vega, J. C., Eguro, K., Koch, D., Handagala, S., Leeser, M., Herbordt, M., Shahzad, H., Hofste, P., Ringlein, B., Szefer, J., Sanaullah, A., & Tessier, R.
- Affiliations
- University of Florida (US)
- University of Toronto (CA)
- Microsoft (United States) (US)
- University of Manchester (GB)
- Northeastern University (US)
- Boston University (US)
- IBM (United States) (US)
- IBM Research - Zurich (CH)
- Yale University (US)
- Red Hat (United States) (US)
- University of Massachusetts Amherst (US)
- ORCID
- Venue
- ACM Transactions on Reconfigurable Technology and Systems
- Published
- 2022-02-04
- Volume
- 15
- Issue
- 3
- Pages
- 1-42
- ISSN
- 1936-7406 (ISSN-L)
- 1936-7414
- OpenAlex
- W4210258659
- Cited by
- 158
- References
- 143
- Citations by year
- 2026: 21 · 2025: 43 · 2024: 43 · 2023: 40 · 2022: 11
- FWCI
- 26.12
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- ACM Transactions on Reconfigurable Technology and Systems (publishedVersion)
- Topics
- Physical Sciences › Computer Science › Hardware and Architecture › Embedded Systems Design Techniques
- Physical Sciences › Computer Science › Information Systems › Cloud Computing and Resource Management
- Physical Sciences › Engineering › Electrical and Electronic Engineering › VLSI and FPGA Design Techniques
- Keywords
- FPGA acceleration, reconfigurable computing, data centers, cloud computing, scalability, virtualization, middleware, operating system abstractions, hardware security, tenant isolation, attack scenarios, deep learning acceleration, scientific computing, FPGA architecture
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.