Zhaohua Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhaohua Wang has a total of 20 co-authors. Here are the top co-authors by number of shared publications:
- Zhenyu Li
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China and University of Chinese Academy of Sciences, Beijing, China
6 shared publications - Wenhao Wu
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
4 shared publications - Qinxin Li
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
3 shared publications - Zihan Li
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
3 shared publications - Chuan Gao
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
2 shared publications - Jin Yan
China Internet Network Information Center, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 12 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 604-619
Taming DNS Responses: A Measurement Study of Response Size and Its Impact
Modern DNS has evolved beyond traditional name resolution, increasingly serving as a carrier for structured metadata and service-discovery information. This functional enrichment inevitably drives sustained growth in DNS response payloads. While essential for modern ecosystem features, this growth is also making DNS response size the center of amplification risk. Despite its importance, the behavior of response size in the wild remains poorly understood, and we still lack a global, systematic view to guide its configuration and control. To this end, we conduct a year-long measurement of response-size behavior across about one million open DNS resolvers. We characterize the UDP response-size limits enforced by these resolvers and find that they cluster around a small set of common values. We also demonstrate how configuration changes at major public resolvers propagate across the DNS ecosystem. We further show that regional differences in resolver response-size limits lead to uneven amplification risk and expose fundamental limitations of existing static and upstream-based defenses. This complex response behavior stems from the tension between mitigating amplification risk through small fixed response-size limits and supporting diverse applications that require large responses. To address this tension, we propose Adaptive Response Scaling (ARS), a lightweight dynamic response-control mechanism that reduces amplification by over 70% while preserving DNS functionality.
- 12 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 604-619
Open AccessTaming DNS Responses: A Measurement Study of Response Size and Its Impact
Modern DNS has evolved beyond traditional name resolution, increasingly serving as a carrier for structured metadata and service-discovery information. This functional enrichment inevitably drives sustained growth in DNS response payloads. While essential for modern ecosystem features, this growth is also making DNS response size the center of amplification risk. Despite its importance, the behavior of response size in the wild remains poorly understood, and we still lack a global, systematic view to guide its configuration and control. To this end, we conduct a year-long measurement of response-size behavior across about one million open DNS resolvers. We characterize the UDP response-size limits enforced by these resolvers and find that they cluster around a small set of common values. We also demonstrate how configuration changes at major public resolvers propagate across the DNS ecosystem. We further show that regional differences in resolver response-size limits lead to uneven amplification risk and expose fundamental limitations of existing static and upstream-based defenses. This complex response behavior stems from the tension between mitigating amplification risk through small fixed response-size limits and supporting diverse applications that require large responses. To address this tension, we propose Adaptive Response Scaling (ARS), a lightweight dynamic response-control mechanism that reduces amplification by over 70% while preserving DNS functionality.
- Open Access
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2995-3005
Tracking the Stray Sheep: Understanding DNS Response Manipulation in the Wild
- 12 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 2995-3005
Open AccessTracking the Stray Sheep: Understanding DNS Response Manipulation in the Wild
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (5) , 1-17
Ppin: Anomaly-based Intrusion Detection on Provenance Graph via Behavior Pattern Mining
- 1 January 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Dependable and Secure Computing
- Vol. 23 (5) , 1-17
Ppin: Anomaly-based Intrusion Detection on Provenance Graph via Behavior Pattern Mining
- Open Access
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 4745-4754
ODNS Clustering: Unveiling Client-Side Dependency in Open DNS Infrastructure
- 22 April 2025
- Published by Association for Computing Machinery (ACM)
- p. 4745-4754
Open AccessODNS Clustering: Unveiling Client-Side Dependency in Open DNS Infrastructure
- Open Access
- 3 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 142-148
SAROS: A Self-Adaptive Routing Oblivious Sampling Method for Network-wide Heavy Hitter Detection
Network-wide heavy hitter detection is usually performed by sampling on several network measurement points (NMPs) and merging the measurement results in the centralized controller to get a network-wide view. However, a packet may pass several NMPs and be counted multiple times when measurement results are merged, which causes the double-counting problem and leads to incorrect detection. Existing studies either overlook this problem or require significant memory usage. This paper proposes SAROS, a self-adaptive routing oblivious sampling method for accurate network-wide heavy hitter detection. Specifically, SAROS exploits a sampling mechanism in the data plane, where the sampling threshold on each measurement point is predicted and adaptively set by the control plane. Such guidance from the control plane greatly reduces the memory usage in the data plane, while mitigating the double-counting problem. Experimental results show that, compared with existing solutions, SAROS improves the F1-Score of heavy hitter detection by 10 ∼ 40%.
- 3 August 2024
- Published by Association for Computing Machinery (ACM)
- p. 142-148
Open AccessSAROS: A Self-Adaptive Routing Oblivious Sampling Method for Network-wide Heavy Hitter Detection
Network-wide heavy hitter detection is usually performed by sampling on several network measurement points (NMPs) and merging the measurement results in the centralized controller to get a network-wide view. However, a packet may pass several NMPs and be counted multiple times when measurement results are merged, which causes the double-counting problem and leads to incorrect detection. Existing studies either overlook this problem or require significant memory usage. This paper proposes SAROS, a self-adaptive routing oblivious sampling method for accurate network-wide heavy hitter detection. Specifically, SAROS exploits a sampling mechanism in the data plane, where the sampling threshold on each measurement point is predicted and adaptively set by the control plane. Such guidance from the control plane greatly reduces the memory usage in the data plane, while mitigating the double-counting problem. Experimental results show that, compared with existing solutions, SAROS improves the F1-Score of heavy hitter detection by 10 ∼ 40%.
- 15 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (5) , 1390-1405
Large-Scale Measurements and Prediction of DC-WAN Traffic
Large cloud service providers have built an increasing number of geo-distributed data centers (DCs) connected by Wide Area Networks (WANs). These DC-WANs carry both high-priority traffic from interactive services and low-priority traffic from bulk transfers. Given that a DC-WAN is an expensive resource, providers often manage it via traffic engineering algorithms that rely on accurate predictions of inter-DC high-priority (delay-sensitive) traffic. In this article, we perform a large-scale measurement study of high-priority inter-DC traffic from Baidu. We measure how inter-DC traffic varies across their global DC-WAN and show that most existing traffic prediction methods either cannot capture the complex traffic dynamics or overlook traffic interrelations among DCs. Building on our measurements, we propose the In terrelated- Te mporal G raph Convolutional Net work (IntegNet) model for inter-DC traffic prediction. In contrast to prior efforts, our model exploits both temporal traffic patterns and inferred co-dependencies between DC pairs. IntegNet forecasts the capacity needed for high-priority traffic demands by accounting for the balance between resource provisioning (i.e., allocating resources exceeding actual demand) and QoS losses (i.e., allocating fewer resources than actual demand). Our experiments show that IntegNet can keep a very limited QoS loss, while also reducing overprovisioning by up to 42.1% compared to the state-of-the-art and up to 66.2% compared to the traditional method used in DC-WAN traffic engineering.
- 15 February 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Parallel and Distributed Systems
- Vol. 34 (5) , 1390-1405
Large-Scale Measurements and Prediction of DC-WAN Traffic
Large cloud service providers have built an increasing number of geo-distributed data centers (DCs) connected by Wide Area Networks (WANs). These DC-WANs carry both high-priority traffic from interactive services and low-priority traffic from bulk transfers. Given that a DC-WAN is an expensive resource, providers often manage it via traffic engineering algorithms that rely on accurate predictions of inter-DC high-priority (delay-sensitive) traffic. In this article, we perform a large-scale measurement study of high-priority inter-DC traffic from Baidu. We measure how inter-DC traffic varies across their global DC-WAN and show that most existing traffic prediction methods either cannot capture the complex traffic dynamics or overlook traffic interrelations among DCs. Building on our measurements, we propose the In terrelated- Te mporal G raph Convolutional Net work (IntegNet) model for inter-DC traffic prediction. In contrast to prior efforts, our model exploits both temporal traffic patterns and inferred co-dependencies between DC pairs. IntegNet forecasts the capacity needed for high-priority traffic demands by accounting for the balance between resource provisioning (i.e., allocating resources exceeding actual demand) and QoS losses (i.e., allocating fewer resources than actual demand). Our experiments show that IntegNet can keep a very limited QoS loss, while also reducing overprovisioning by up to 42.1% compared to the state-of-the-art and up to 66.2% compared to the traditional method used in DC-WAN traffic engineering.