My name is Gangmuk Lim. I am a 2nd 3rd year Ph.D. student at the Computer Science Department of the University of Illinois Urbana-Champaign, working with Professor Brighten Godfrey. Before starting my Ph.D., I earned my Bachelor and Master at the Computer Science at UNIST, South Korea, working with Professor Myeongjae Jeon.
My research primarily revolves around improving performance of microservice application in the cloud, reliability of cluster management system, and performance & resource utilization for machine learning applications.
The most recent research project focuses on designing new service mesh system to minimize end-to-end latency and monetary costs associated with large scale microservice applications spanning geo-distributed multiple clusters in the cloud.
Previously, I also worked on building a resource efficient machine learning training system with smart GPU sharing (co-locating multiple DNN training jobs within a GPU) and improves throughput with higher GPU resource utilization.
You can find my full CV
here.
Most likely, you can find me at Siebel building on 3rd floor at Urbana, IL.
LinkedIn, GitHub
News!
- Opportunities and Challenges in Service Layer Traffic Engineering accepted to HotNets 24'. paper
- Kivi accepted to USENIX ATC 24'. paper
- Gave a talk at KubeCon 23' about Kivi. talk
- Gave a talk at KubeCon 23' about SLATE. talk
- Gave a talk at IstioCon 23' about SLATE.
Publication
- Opportunities and Challenges in Service Layer Traffic Engineering
Gangmuk Lim, Aditya Prerepa, Brighten Godfrey, Radhika Mittal
HotNets 2024 - Kivi: Verification for Cluster Management
Bingzhe Liu, Gangmuk Lim, Ryan Beckett, P. Brighten Godfrey.
USENIX Annual Technical Conference, 2024 - Zico: Efficient GPU Memory Sharing for Concurrent DNN Training
Gangmuk Lim, Jeongseob Ahn, Wencong Xiao, Youngjin Kwon, Myeongjae Jeon.
USENIX Annual Technical Conference, 2021 - Approximate Quantiles for Data Center Telemetry Monitoring
Gangmuk Lim, Myeongjae Jeon, Stavros Volos, Mohamed Hassan, Ze Jin.
IEEE ICDE 2020 short