Carol Hsu has worked in computer systems for years. Her research began with HPC and virtualization at National Tsing Hua University (advised by Jerry Chou), focusing on deploying and running HPC workloads in the cloud. Later, as an engineer in the industry, she worked on public cloud service backends, specializing in Linux containers, microservices, and Kubernetes.

She completed her PhD at Georgia Institute of Technology (advised by Ada Gavrilovska), expanding her system studies into edge computing on 5G networks. Her work explores efficient, fast, and observable system support for edge workloads.

Carol now works as a research engineer in EMEA area.

Selected Publication

Colibri: Efficient Collection of Fine-Grained Resource Metrics Necessary for Mobile Edge Computing, Ke-Jou Hsu, Ketan Bhardwaj, Ada Gavrilovska, The Ninth ACM/IEEE Symposium on Edge Computing (SEC'24)

ShapeShifter: Resolving the Hidden Latency Contention Problem in MEC, Valentin Rakovic, Ke-Jou Hsu, Ketan Bhardwaj, Ada Gavrilovska, Liljana Gavrilovska, The Seventh ACM/IEEE Symposium on Edge Computing (SEC'22)

Performance Benchmarking and Auto-tuning for Scientific Applications on Virtual Cluster, Ke-Jou Hsu, Jerry Chi-Yuan Chou, The Journal of Supercomputing 2021, Springer

DNS Does Not Suffice for MEC-CDN, Ke-Jou Hsu, James Choncholas, Ketan Bhardwaj, Ada Gavrilovska, The Nineteenth ACM Workshop on Hot Topics in Networks (HotNet'20)

Couper: DNN Model Slicing for Visual Analytics Containers at the Edge, Ke-Jou Hsu, Ketan Bhardwaj, Ada Gavrilovska, The Fourth ACM/IEEE Symposium on Edge Computing (SEC'19)

Kubernetes Cookbook - 2nd edition, Hideto Saito, Hui-Chuan Chloe Lee, Ke-Jou Carol Hsu, Packt Publishing 05/2018 (The same team published 1st edition at 06/2015)

Selected Presentation

Fine-grained, Lightweight Resource Metric Collection in Mobile Edge Computing – Alumni Seminar, NTHU (09/2023)

MEC-in-a-Box – Demo, Application Driven Architecture (ADA) Center Annual Symposium, SRC (05/2020)