Zedong Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zedong Liu has a total of 28 co-authors. Here are the top co-authors by number of shared publications:
- Dingwen Tao
Chinese Academy of Sciences
4 shared publications - Guangming Tan
Chinese Academy of Sciences
4 shared publications - Wenjing Huang
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
4 shared publications - Yida Gu
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
4 shared publications - Hairui Zhao
Jilin University
3 shared publications - Bing Lu
Institute of Computing Technology, Chinese Academy of Sciences, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-15
KVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving
LLMs are widely adopted in production, pushing inference systems to their limits. Disaggregated LLM serving (e.g., PD separation and KV state disaggregation) improves scalability and cost efficiency, but it also turns KV into an explicit payload crossing network and storage boundaries, making KV a dominant end-to-end bottle-neck. Existing KV compression are typically statically configured at runtime, despite production service context varies over time in workload mix, bandwidth, and SLO/quality budgets. As a result, a fixed choice can be suboptimal or even increase latency. We present KVServe, the first service-aware and adaptive KV communication compression framework for disaggregated LLM serving: KVServe (1) unifies KV compression into a modular strategy space with new components and cross-method recomposition; (2) introduces Bayesian Profiling Engine that efficiently searches this space and distills a 3D Pareto candidate set, reducing 50x offline search overhead; and (3) deploys a Service-Aware Online Controller that combines an analytical latency model with a lightweight bandit to select profiles under constraints and correct offline-to-online mismatch. Integrated into vLLM and evaluated across datasets, models, GPUs and networks, KVServe achieves up to 9.13x JCT speedup in PD-separated serving and up to 32.8x TTFT reduction in KV-disaggregated serving.
- 11 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 1-15
Open AccessKVServe: Service-Aware KV Cache Compression for Communication-Efficient Disaggregated LLM Serving
LLMs are widely adopted in production, pushing inference systems to their limits. Disaggregated LLM serving (e.g., PD separation and KV state disaggregation) improves scalability and cost efficiency, but it also turns KV into an explicit payload crossing network and storage boundaries, making KV a dominant end-to-end bottle-neck. Existing KV compression are typically statically configured at runtime, despite production service context varies over time in workload mix, bandwidth, and SLO/quality budgets. As a result, a fixed choice can be suboptimal or even increase latency. We present KVServe, the first service-aware and adaptive KV communication compression framework for disaggregated LLM serving: KVServe (1) unifies KV compression into a modular strategy space with new components and cross-method recomposition; (2) introduces Bayesian Profiling Engine that efficiently searches this space and distills a 3D Pareto candidate set, reducing 50x offline search overhead; and (3) deploys a Service-Aware Online Controller that combines an analytical latency model with a lightweight bandit to select profiles under constraints and correct offline-to-online mismatch. Integrated into vLLM and evaluated across datasets, models, GPUs and networks, KVServe achieves up to 9.13x JCT speedup in PD-separated serving and up to 32.8x TTFT reduction in KV-disaggregated serving.
- Open Access
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 164-176
PRISM: An Efficient GPU-Based Lossy Compression Framework for Progressive Data Retrieval with Multi-Level Interpolation
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 164-176
Open AccessPRISM: An Efficient GPU-Based Lossy Compression Framework for Progressive Data Retrieval with Multi-Level Interpolation
- Open Access
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 425-438
CCL-D: A High-Precision Diagnostic System for Slow and Hang Anomalies in Large-Scale Model Training
- 28 January 2026
- Published by Association for Computing Machinery (ACM)
- p. 425-438
Open AccessCCL-D: A High-Precision Diagnostic System for Slow and Hang Anomalies in Large-Scale Model Training
- Open Access
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1299-1314
MANS: Efficient and Portable ANS Encoding for Multi-Byte Integer Data on CPUs and GPUs
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 1299-1314
Open AccessMANS: Efficient and Portable ANS Encoding for Multi-Byte Integer Data on CPUs and GPUs