| Research | 研究工作 |
- Performance Improvement of Service Mesh
|
|
2021 - 2022 |
| - We analyzed the architecture of Istio and found the performance bottleneck causing the high resource consumption. |
| - We proposed an adaptive configuration loading mechanism, reducing memory consumption of data plane by 98%. |
| Service Mesh Istio Microservice Performance Improvement Production-ready |
- Operation Enhancement of Service Mesh
|
|
2020 - 2021 |
| - We designed the escape solution for Envoy based on Iptables, which enables the system recover from the failure |
| quickly and improves the efficiency of troubleshooting. |
| - We analyzed the architecture of Envoy and designed an expert system to determine the root cause by extending |
| and collecting the accesslogs in a low-intrusive manner, which improves the efficiency operation and maintenance. |
| Service Mesh Envoy Microservice Ops Production-ready |
- Security Review of Serverless Computing: Challenges, Solutions, and Opportunities
|
2020 - 2022 |
| - We introduced the security challenges of serverless computing and compared the solutions adopted in the academia, |
| industry, and open-sourced platforms. |
| - We also analyzed the gap between solutions and proposed the potential research opportunities. |
| Survey Serverless Security |
- PROPHET: Efficient and Intelligent Orchestrator for Microservices Scheduling and Scaling
|
2019 - 2020 |
| - We proposed a ranking-based p-batch scheduling mechanism, which employs a pairwise ranker to achieve
rapid |
| and resource-efficient deployment for large-scale microservices to improve the resource utilization by reducing |
| the number of running nodes in the cluster. |
| - We proposed a proactive prediction-based scaling mechanism, which automatically scales microservices in advance |
| based on the resource usage prediction. This scaling mechanism can effectively alleviate the sluggishness in scaling |
| and mitigate resource contention and service interruption. |
| Microservice Kubernetes Scheduling & Scaling AI |
- SDNKeeper: Lightweight Resource Protection and Management System for SDN-based Cloud
|
2017 - 2019 |
| - We proposed a policy-based fine-grained access control mechanism, which can effectively prevent unauthorized |
| access requests initialed by users and malicious tampering of controllers and network resources by filtering and |
| checking requests with predefined policies. |
| SDN OpenDaylight Access Control |
- Fault Management in Software-Defined Networking: A Survey
|
2017 - 2018 |
| - We first categorized the solutions according to architecture of SDN. |
| - Then, we analyzed and summarized thefault management solutions of SDN from the aspects of fault analysis, |
| fault detection, fault diagnosis,fault failover, and fault tolerance. |
| - At last, we proposed the potential research opportunities. |
| Survey SDN Fault Management |
- Development of SDN controller for Cloud Data Center
|
2016 |
| - We developed the SDN controller based on the Open SDN Controller of Cisco. |
| - We chose the Tencent Cloud Data Center as the test bed to perform the network configuration test. |
| Intern SDN OpenDaylight |
- RuleScope: Inspecting Forwarding Faults for Software-Defined Networking
|
2015 - 2016 |
| - We presented RuleScope, a more comprehensive solution for inspecting SDN forwarding. |
| - RuleScope offers a series of accurate and efficient algorithms for detecting and troubleshooting rule faults. |
| They inspect forwarding behavior using customized probe packets to exercise data-plane rules. |
| - The detection algorithm exposes not only missing faults but also priority faults. |
| - Beyond simply detecting rule faults, the troubleshooting algorithms uncover actual dataplane flow tables. |
| They help track real-time forwarding status and benefit reliable network monitoring. |
| SDN Configuration Consistency |
