Liang Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Liang Wang has a total of 74 co-authors. Here are the top co-authors by number of shared publications:
- Chao Pei
Tencent, Beijing, China
5 shared publications - Jian Wang
Tencent, Beijing, China
5 shared publications - Jian Zhao
Tencent, Beijing, China
5 shared publications - Kai Ren
Tencent, Beijing, China
4 shared publications - Congcong Miao
National University of Singapore, Singapore, Singapore
3 shared publications - Heng Yu
Tsinghua University, Beijing, China
3 shared publications
Latest Publications
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 535-549
Pegasus: A Data Center Network for Bare-Metal AI Cloud
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 535-549
Open AccessPegasus: A Data Center Network for Bare-Metal AI Cloud
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 433-446
Dorado: Scaling SmartNIC Session Tables on Commodity DDRs
FPGA-based SmartNICs are widely deployed for cloud network function acceleration, but their memory subsystem is under increasing pressure because of large session tables. Conventional wisdom suggests that high packet processing performance relies on advanced memories (e.g., SRAM, HBM), but those are costly to add at cloud scale. Dorado is a novel design that scales SmartNIC session tables entirely on inexpensive DDR modules. At the heart of Dorado are three new techniques that extract commodity DDR performance by restructuring session table layout, decomposing processing pipelines to reduce locking, and scheduling memory accesses to minimize stalls. Our testbed results show that Dorado improves packet processing rates by 33%, even with fewer hardware resources. Further, we have deployed Dorado to millions of servers, processing network traffic from billions of users on a large public cloud for over three years. Our production results show that Dorado can accommodate up to 16M session entries, reduce memory cost by 80%, while enabling 50Mpps line-rate processing.
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 433-446
Open AccessDorado: Scaling SmartNIC Session Tables on Commodity DDRs
FPGA-based SmartNICs are widely deployed for cloud network function acceleration, but their memory subsystem is under increasing pressure because of large session tables. Conventional wisdom suggests that high packet processing performance relies on advanced memories (e.g., SRAM, HBM), but those are costly to add at cloud scale. Dorado is a novel design that scales SmartNIC session tables entirely on inexpensive DDR modules. At the heart of Dorado are three new techniques that extract commodity DDR performance by restructuring session table layout, decomposing processing pipelines to reduce locking, and scheduling memory accesses to minimize stalls. Our testbed results show that Dorado improves packet processing rates by 33%, even with fewer hardware resources. Further, we have deployed Dorado to millions of servers, processing network traffic from billions of users on a large public cloud for over three years. Our production results show that Dorado can accommodate up to 16M session entries, reduce memory cost by 80%, while enabling 50Mpps line-rate processing.
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 447-460
XFir: Accelerating New-Flow Setup on Host Servers of a Large Cloud Network
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 447-460
Open AccessXFir: Accelerating New-Flow Setup on Host Servers of a Large Cloud Network
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 519-534
DistDPU: A Disaggregated DPU Architecture for High-Performance and Cost-Efficient AI Clouds
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 519-534
Open AccessDistDPU: A Disaggregated DPU Architecture for High-Performance and Cost-Efficient AI Clouds
- Open Access
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 663-676
Fornax: A Hardware-Centric Session Management in Large Public Cloud Network
- 27 August 2025
- Published by Association for Computing Machinery (ACM)
- p. 663-676
Open AccessFornax: A Hardware-Centric Session Management in Large Public Cloud Network
- Open Access
- 18 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 112-120
Multi-agent Multi-game Entity Transformer: Towards Generalist Models in MARL
- 18 December 2024
- Published by Association for Computing Machinery (ACM)
- p. 112-120
Open AccessMulti-agent Multi-game Entity Transformer: Towards Generalist Models in MARL