Ying Wan
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Ying Wan has a total of 71 co-authors. Here are the top co-authors by number of shared publications:
Bin Liu
Tsinghua University,China
13 shared publicationsChuwen Zhang
Tsinghua University
9 shared publicationsHaoyu Song
9 shared publicationsYang Xu
Fudan University,School of Computer Science,China
8 shared publicationsTian Pan
7 shared publicationsYi Wang
7 shared publications
Latest Publications
- Open Access
- 15 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (18) , 4181
From Inference to Attack: AI-Driven BFI Forgery Against Unknown MU-MIMO Schedulers via TD-MoE
Compressed beamforming feedback information (BFI) reduces feedback overhead in multi-user multiple-input multiple-output (MU-MIMO) systems, but it also exposes the feedback signal that drives user scheduling and precoding. This paper studies whether an external observer can learn an unknown BFI-based scheduler and then use the learned behavior to guide feedback forgery in a protocol-informed physical-layer MU-MIMO simulation. We propose Trajectory Decoding Mixture-of-Experts (TD-MoE), an AI-driven scheduler-inference model that predicts selected user subsets through progressive decoding and latent expert routing, without scheduler-identity labels. We then design a full-feedback BFI forgery method that combines feedback-geometry-guided finite perturbation search with a causal feedback-only attack-or-abstain policy for selecting attack-worthy feedback instances. The attack perturbs scheduler decisions and the feedback-side regularized zero-forcing (regularized-ZF) geometry. In the mixed-five-scheduler inference benchmark, TD-MoE achieves an average intersection over union (IoU) of 0.740 and a Top-1 exact-match accuracy of 0.543, outperforming baselines without trajectory supervision or expert specialization. Under online full-feedback forgery, blind all-attack and random triggering produce limited average degradation, whereas feedback-side selective triggering substantially improves large-drop targeting. In particular, under the default rescheduled full-feedback setting, the learned attack-or-abstain policy achieves an attacked-only mean drop of , a median drop of , and attack success rates of and at the 10% and 20% degradation thresholds, denoted by and , respectively, without using true-channel measurements, achieved-rate feedback, future samples, or test-set sorting during online attack selection. These results show that, within this protocol-informed physical-layer simulation setting, BFI-based resource scheduling can be inferred from compressed feedback and that scheduler confidentiality alone does not protect feedback-driven wireless control loops.- 15 September 2026
- Published by MDPI AG in Electronics
- Vol. 15 (18) , 4181
Open AccessFrom Inference to Attack: AI-Driven BFI Forgery Against Unknown MU-MIMO Schedulers via TD-MoE
Compressed beamforming feedback information (BFI) reduces feedback overhead in multi-user multiple-input multiple-output (MU-MIMO) systems, but it also exposes the feedback signal that drives user scheduling and precoding. This paper studies whether an external observer can learn an unknown BFI-based scheduler and then use the learned behavior to guide feedback forgery in a protocol-informed physical-layer MU-MIMO simulation. We propose Trajectory Decoding Mixture-of-Experts (TD-MoE), an AI-driven scheduler-inference model that predicts selected user subsets through progressive decoding and latent expert routing, without scheduler-identity labels. We then design a full-feedback BFI forgery method that combines feedback-geometry-guided finite perturbation search with a causal feedback-only attack-or-abstain policy for selecting attack-worthy feedback instances. The attack perturbs scheduler decisions and the feedback-side regularized zero-forcing (regularized-ZF) geometry. In the mixed-five-scheduler inference benchmark, TD-MoE achieves an average intersection over union (IoU) of 0.740 and a Top-1 exact-match accuracy of 0.543, outperforming baselines without trajectory supervision or expert specialization. Under online full-feedback forgery, blind all-attack and random triggering produce limited average degradation, whereas feedback-side selective triggering substantially improves large-drop targeting. In particular, under the default rescheduled full-feedback setting, the learned attack-or-abstain policy achieves an attacked-only mean drop of , a median drop of , and attack success rates of and at the 10% and 20% degradation thresholds, denoted by and , respectively, without using true-channel measurements, achieved-rate feedback, future samples, or test-set sorting during online attack selection. These results show that, within this protocol-informed physical-layer simulation setting, BFI-based resource scheduling can be inferred from compressed feedback and that scheduler confidentiality alone does not protect feedback-driven wireless control loops. - Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 334-352
Turbo: Efficiently Serving Long-Context Large Language Models with In-Network Aggregation
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 334-352
Open AccessTurbo: Efficiently Serving Long-Context Large Language Models with In-Network Aggregation
- Open Access
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 288-299
MegaTurbo: A Scalable FPGA-based Engine for MegaFlow Classifier in Open vSwitch
Open vSwitch (OVS) is a key component in cloud and data center networks, yet its MegaFlow classifier imposes significant CPU overhead. Existing SmartNIC-based acceleration approaches for the MegaFlow classifier typically employ simplistic hardware offloading techniques, which exhibit limited scalability for dynamic, large-scale flow tables. Motivated by these challenges, we argue that a hardware accelerator specifically tailored for the MegaFlow classifier is necessary, forming the basis of our FPGA-based solution, MegaTurbo. The core innovations of MegaTurbo are threefold: (1) a scalable and hardware-friendly decision-tree based packet classification algorithm, specifically optimized for the structure of MegaFlow rules; (2) a novel hardware architecture incorporating multiple pipelined matching engines, designed to process multiple decision trees generated by the software algorithm in parallel; and (3) a heterogeneous framework composed of CPU and FPGA, which can work together to support online rule updates, with little and bounded impact on rule searching. Experimental results on a Xilinx Virtex UltraScale+ FPGA demonstrate that MegaTurbo achieves a sustained classification throughput of 500 MPPS while supporting dynamic rule updates at 300-500 KUPS on 100K-scale rulesets. These results not only validate the effectiveness of our domain-specific co-design approach, but also highlight the potential of FPGA-based SmartNICs to address the performance bottlenecks of software switches in large-scale cloud and data center networks.
- 21 February 2026
- Published by Association for Computing Machinery (ACM)
- p. 288-299
Open AccessMegaTurbo: A Scalable FPGA-based Engine for MegaFlow Classifier in Open vSwitch
Open vSwitch (OVS) is a key component in cloud and data center networks, yet its MegaFlow classifier imposes significant CPU overhead. Existing SmartNIC-based acceleration approaches for the MegaFlow classifier typically employ simplistic hardware offloading techniques, which exhibit limited scalability for dynamic, large-scale flow tables. Motivated by these challenges, we argue that a hardware accelerator specifically tailored for the MegaFlow classifier is necessary, forming the basis of our FPGA-based solution, MegaTurbo. The core innovations of MegaTurbo are threefold: (1) a scalable and hardware-friendly decision-tree based packet classification algorithm, specifically optimized for the structure of MegaFlow rules; (2) a novel hardware architecture incorporating multiple pipelined matching engines, designed to process multiple decision trees generated by the software algorithm in parallel; and (3) a heterogeneous framework composed of CPU and FPGA, which can work together to support online rule updates, with little and bounded impact on rule searching. Experimental results on a Xilinx Virtex UltraScale+ FPGA demonstrate that MegaTurbo achieves a sustained classification throughput of 500 MPPS while supporting dynamic rule updates at 300-500 KUPS on 100K-scale rulesets. These results not only validate the effectiveness of our domain-specific co-design approach, but also highlight the potential of FPGA-based SmartNICs to address the performance bottlenecks of software switches in large-scale cloud and data center networks.
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 5985-5998
LAPS: Latency Aware Packet Spraying on Unequal-Cost Multi-Path Data Center Networks
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Networking
- Vol. 34, 5985-5998
LAPS: Latency Aware Packet Spraying on Unequal-Cost Multi-Path Data Center Networks
- Open Access
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 11-18
LAPS: Joint Load Balancing and Congestion Control on Unequal-cost Multi-path Data Center Networks
- 8 September 2025
- Published by Association for Computing Machinery (ACM)
- p. 11-18
Open AccessLAPS: Joint Load Balancing and Congestion Control on Unequal-cost Multi-path Data Center Networks
- 9 December 2024
- Published by Elsevier in Computer Networks
- Vol. 257
DeFlow: Differential flowlet switching for load balancing in datacenter networks
- 9 December 2024
- Published by Elsevier in Computer Networks
- Vol. 257
DeFlow: Differential flowlet switching for load balancing in datacenter networks
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 32 (6) , 4842-4854
OBMA: Scalable Route Lookups With Fast and Zero-Interrupt Updates
Software-based IP route lookup is a key component for packet forwarding in Software Defined Networks. Running lookup algorithms on commodity CPUs is flexible and scalable, which shows advantages on cost and power consumption over the hardware-based forwarding engines. However, dynamic network functions and services make route updates more frequent than ever. Existing algorithms often fall short of the incremental update requirements. In this paper, we propose the Overlay BitMap Algorithm (OBMA), which contains several variations, to support extraordinary update performance while maintaining the highest-in-class lookup speed and storage efficiency. Starting from the basic OBMA_B, we develop two variations with different tradeoffs for different application scenarios. OBMA_L supports faster lookups than OBMA_B at a small cost of update speed. OBMA_S achieves better storage efficiency than OBMA_B at a small cost of lookup throughput. We run our algorithms on a commodity CPU and evaluate them with real-world route tables and traces. The experiments show that OBMA achieves the lowest memory footprint, the highest update speed, and over 200 Mpps lookup throughput. Specifically, OBMA_S reduces the memory footprint to 3.98 bytes/prefix which is 25.33% smaller that of the state-of-the-art Poptrie; OBMA_L supports 252.02 Mpps lookup throughput with a single thread, and more than 600 Mpps with multiple parallel threads in a single CPU, significantly outperforming the state-of-the-art Poptrie and SAIL; OBMA_B supports updates at a rate of 14.58M updates/s which is 15 times faster than Poptrie. The tests show that the update process has little interference with the lookup process for OBMA, and achieves zero-interrupt to lookups with multiple threads.
- 11 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE/ACM Transactions on Networking
- Vol. 32 (6) , 4842-4854
OBMA: Scalable Route Lookups With Fast and Zero-Interrupt Updates
Software-based IP route lookup is a key component for packet forwarding in Software Defined Networks. Running lookup algorithms on commodity CPUs is flexible and scalable, which shows advantages on cost and power consumption over the hardware-based forwarding engines. However, dynamic network functions and services make route updates more frequent than ever. Existing algorithms often fall short of the incremental update requirements. In this paper, we propose the Overlay BitMap Algorithm (OBMA), which contains several variations, to support extraordinary update performance while maintaining the highest-in-class lookup speed and storage efficiency. Starting from the basic OBMA_B, we develop two variations with different tradeoffs for different application scenarios. OBMA_L supports faster lookups than OBMA_B at a small cost of update speed. OBMA_S achieves better storage efficiency than OBMA_B at a small cost of lookup throughput. We run our algorithms on a commodity CPU and evaluate them with real-world route tables and traces. The experiments show that OBMA achieves the lowest memory footprint, the highest update speed, and over 200 Mpps lookup throughput. Specifically, OBMA_S reduces the memory footprint to 3.98 bytes/prefix which is 25.33% smaller that of the state-of-the-art Poptrie; OBMA_L supports 252.02 Mpps lookup throughput with a single thread, and more than 600 Mpps with multiple parallel threads in a single CPU, significantly outperforming the state-of-the-art Poptrie and SAIL; OBMA_B supports updates at a rate of 14.58M updates/s which is 15 times faster than Poptrie. The tests show that the update process has little interference with the lookup process for OBMA, and achieves zero-interrupt to lookups with multiple threads.
- Open Access
- 23 August 2024
- Published by Elsevier in SSRN Electronic Journal
Deflow: Differential Flowlet Switching for Load Balancing in Datacenter Networks
- 23 August 2024
- Published by Elsevier in SSRN Electronic Journal
Open AccessDeflow: Differential Flowlet Switching for Load Balancing in Datacenter Networks
- 9 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 252-258
Multi-Stage Flow Table Caching: From Theory to Algorithm
Flow table capacity in programmable switches is constrained due to the limited on-chip hardware resource. The current mainstream approach is to cache only the popular rules in hardware. By taking advantage of traffic locality, the majority of packets can be forwarded directly after matching the rules cached in hardware and the remaining missed packets are handled by software that accommodates the full flow table. Existing works focus on selecting the cache entries for a single-stage flow table to achieve a high cache hit-rate, which cannot adapt to multi-stage flow tables. Due to hardware constraints as well as service requirements, it is often necessary to decompose a single-stage flow table to a multi-stage flow table or directly create multiple stages of tables in hardware. For the first time, we abstract and model the multi-stage flow table caching problem and prove the NP-hardness of the Optimal Multi-stage Flow table Caching (OMFC). Further, we propose a Greedy Caching Algorithm (GCA) for OMFC, which considers both the rule popularity across multiple stages of flow tables and entry popularity within the same stage of flow table when determining the content and size of the multi-stage flow tables. The simulation results show that GCA achieves a l0~30% higher cache hit-rate than the existing algorithms.
- 9 July 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 252-258
Multi-Stage Flow Table Caching: From Theory to Algorithm
Flow table capacity in programmable switches is constrained due to the limited on-chip hardware resource. The current mainstream approach is to cache only the popular rules in hardware. By taking advantage of traffic locality, the majority of packets can be forwarded directly after matching the rules cached in hardware and the remaining missed packets are handled by software that accommodates the full flow table. Existing works focus on selecting the cache entries for a single-stage flow table to achieve a high cache hit-rate, which cannot adapt to multi-stage flow tables. Due to hardware constraints as well as service requirements, it is often necessary to decompose a single-stage flow table to a multi-stage flow table or directly create multiple stages of tables in hardware. For the first time, we abstract and model the multi-stage flow table caching problem and prove the NP-hardness of the Optimal Multi-stage Flow table Caching (OMFC). Further, we propose a Greedy Caching Algorithm (GCA) for OMFC, which considers both the rule popularity across multiple stages of flow tables and entry popularity within the same stage of flow table when determining the content and size of the multi-stage flow tables. The simulation results show that GCA achieves a l0~30% higher cache hit-rate than the existing algorithms.
- 17 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
CoLUE: Collaborative TCAM Update in SDN Switches
With the rapidly changing network, rule update in TCAM has become the bottleneck for application performance. In traditional software-defined networks, some application policies are deployed at the edge switches, while the scarce TCAM spaces exacerbate the frequency and difficulty of rule updates. This paper proposes CoLUE, a framework which groups rules into switches in a balance and dependency minimum way. CoLUE is the first work that combines TCAM update and rule placement, making full use of TCAM in distributed switches. Not only does it accelerate update speed, it also keeps the TCAM space load-balance across switches. Composed of ruleset decomposition and subset distribution, CoLUE has an NP-completeness challenge. We propose heuristic algorithms to calculate a near-optimal rule placement scheme. Our evaluations show that CoLUE effectively balances TCAM space load and reduces the average update cost by more than 1.45 times and the worst-case update cost by up to 5.46 times, respectively.
- 17 May 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-10
CoLUE: Collaborative TCAM Update in SDN Switches
With the rapidly changing network, rule update in TCAM has become the bottleneck for application performance. In traditional software-defined networks, some application policies are deployed at the edge switches, while the scarce TCAM spaces exacerbate the frequency and difficulty of rule updates. This paper proposes CoLUE, a framework which groups rules into switches in a balance and dependency minimum way. CoLUE is the first work that combines TCAM update and rule placement, making full use of TCAM in distributed switches. Not only does it accelerate update speed, it also keeps the TCAM space load-balance across switches. Composed of ruleset decomposition and subset distribution, CoLUE has an NP-completeness challenge. We propose heuristic algorithms to calculate a near-optimal rule placement scheme. Our evaluations show that CoLUE effectively balances TCAM space load and reduces the average update cost by more than 1.45 times and the worst-case update cost by up to 5.46 times, respectively.