The UTNS research group at the University of Texas at Austin works on foundational systems and networking problems that arise in supporting big data analytics, distributed machine learning, and large-scale distributed applications. Our work explores both theoretical underpinnings and system design/implementation, and builds on ideas from adjacent areas such as game theory, optimization, and formal methods.
News
Yeonju Ro has been recognized as an MLCommons ML and Systems Rising Star!
PacketExpress was accepted at SIGCOMM 2026!
Mimesys was accepted at OSDI 2026!
SMEC, UNUM, Syntra, and Symphony were accepted at NSDI 2026!
Rohit Dwivedula, Haoyu Li, and Xiao Zhang won the Amazon AI PhD Fellowship!
PolicySmith was accepted at the HotNets 2025!
Canopy was accepted at EuroSys 2026!
TraceLLM was accepted at EMNLP 2025!
Daehyeok Kim received the NSF CAREER Award!
ConfigBot was accepted at IROS 2025!
PacketExpress was accepted at SIGCOMM 2026!
Mimesys was accepted at OSDI 2026!
SMEC, UNUM, Syntra, and Symphony were accepted at NSDI 2026!
Rohit Dwivedula, Haoyu Li, and Xiao Zhang won the Amazon AI PhD Fellowship!
PolicySmith was accepted at the HotNets 2025!
Canopy was accepted at EuroSys 2026!
TraceLLM was accepted at EMNLP 2025!
Daehyeok Kim received the NSF CAREER Award!
ConfigBot was accepted at IROS 2025!
Featured projects
Featured publications
PacketExpress: Fully Exploiting Large MTUs for Internet Traffic in Private Networks
.
SIGCOMM, 2026.
Mimesys: Generating Realistic Executable Testing Environments from Resource Usage Traces
.
OSDI, 2026.
UNUM: A New Framework for Network Control
.
NSDI, 2026.
Synthesizing a Provably Performant Joint Controller for Low-Latency Video Streaming
.
NSDI, 2026.
SYMPHONY: Improving Memory Management for LLM Inference Workloads
.
NSDI, 2026.
Enabling SLO-Aware 5G Multi-Access Edge Computing with SMEC
.
NSDI, 2026.
Towards Performance Robustness for Microservices
.
NSDI, 2026.
Express Lane to Efficiency and Reliability: Multi-Dimensional Control in Meta’s Express Backbone Network
.
NSDI, 2026.
Canopy: Property-Driven Learning for Congestion Control
.
EuroSys, 2026.