Accepted Posters
The following posters will be presented at the Poster Session and Reception on Tuesday, May 5, 6:00 pm–7:30 pm, in the Lake Washington Ballroom.
Toward an AI-Native Internet: Early Explorations in Agentic Semantic Retrieval
Yixi Chen and Muhammad Bilal, KAUST; Tie Ma, Beihang University & KAUST; Zafar A. Qazi, LUMS & KAUST; Suhaib Fahmy and Marco Canini, KAUST
mrLLM: Fast Multi-Region LLM Inference using Learned Adaptors
Marilyn Rego, University of Michigan; Maxwell Kumbong, Swarthmore College; Hermann Kumbong, Stanford University; Ertza Warraich, Purdue University; Muhammad Shahbaz, University of Michigan
Tundra: A Tunable Communication Abstraction for Distributed Systems
Ethan Lavi, Tianyi Lu, Andrew Li, and Nikos Vasilakis, Brown University
Accelerating Resource Management in Edge Control with Approximate Analytics
Yancheng Yuan, Eric Wang, and Zaoxing Liu, University of Maryland, College Park
Enabling Performant and Flexible Deep Model Inspection
Nengneng Yu, Sixian Xiong, Yibo Zhao, Wei Wang, and Zaoxing Liu, University of Maryland, College Park
Size-aware replication for connection-migration systems
Shuo Li and Michio Honda, University of Edinburgh
Measuring Apex TXT Bloat from DNS-Based Domain Verification: Growth, Privacy Implications, and Design Requirements
Eunbee Hwang, Hyunsoo Kim, and Taekyoung "Ted" Kwon, Seoul National University
Rethinking Network Data Generation: A Study of Tabular Flow-Based Generative Models for Intrusion Detection
Ragini Gupta, University of Illinois at Urbana-Champaign
When Should a Performance Enhancing Proxy Intervene?
Yuan Gao, George Washington University; Shixiong Qi, University of Kentucky; Timothy Wood, George Washington University
Resource-Complementary Scheduling on Datacenter Shared Servers
Yibo Yan, University of Southern California; Zixuan Peng, University of Wisconsin-Madison; Ramesh Govindan and Seo Jin Park, University of Southern California
Quantifying the Cold-Start Tax: Tail-Latency Amplification from Pod Scheduling in Kubernetes Microservice Chains
Sanket Koirala
Towards Extensible, Auditable, and Modular Measurement Specification and Execution in Computer Networking
Innocent Ndubuisi-Obi Jr and Siya Kulkarni, University of Washington
When Server‑Level Load Imbalance Helps: Power Savings without Capacity Loss
Sherif Abdelrazek, Dhyey Thummar, and Ahmed Saeed, Georgia Institute of Technology
Hybrid L7/eBPF Proxy: User-Space Classification with Kernel-Space Socket Splicing
Wen-Ju Chiang, Hsiang-Ming Hung, Yu-Hong Shen, Yen-Ting Kuo, Chien-Chao Tseng, and Meng-Hsun Tsai, National Yang Ming Chiao Tung University
Passive Measurement of Application Latency Under Multiplexing and Encryption
Bhavana Vannarth Shobhana and Srinivas Narayana, Rutgers University
R2CC: Reliable and Resilient Collective Communication
Wei Wang, Nengneng Yu, Sixian Xiong, and Zaoxing Liu, University of Maryland