0xhead

Chin-Jung Hsu

Ph.D Candidate

Computer Science, North Carolina State University

Biography

I am a Ph.D. student, advised by Dr. Vincent W. Freeh, of Computer Science at North Carolina State University, My research primarily focuses on optimizing system performance of large-scale distributed systems. Due to the growing complexity of systems, I am passionate about applying machine learning techniques to address problems in systems such as cloud computing and software-defined storage.

Before starting my Ph.D., I received M.S. in Computer Science from National Tsing Hua University in Taiwan. After graduation, I was an Android engineer at Acer Inc. In 2011, I started my journey in the states. I was a performance measurement intern and a research intern in NetApp for the summer of 2013 and 2014, and in 2015, I interned at AT&T Labs Research, working on software-defined storage.

Interests

Distributed Systems
Storage Systems
System Performance
Artificial Intelligence

Education

Ph.D Candidate in Computer Science, 2018

North Carolina State University
M.S. in Computer Science (Information Systems and Applications), 2009

National Tsing Hua University, Taiwan
B.S. in Computer Science and Information Technology, 2007

Fu Jen Catholic University, Taiwan

Selected Publications

Scout: An Experienced Guide to Find the Best Cloud Configuration

This work proposes an efficient, effective and robust method to find the best configuration in the cloud. SCOUT identifies resource requirements using low-level performance metrics and searches only the spotlight region (configuration space). Our evaluation shows SCOUT is several times better than the state-of-the-art methods.

In USENIX ATC’18 (submitted), 2018.

Low-Level Augmented Bayesian Optimization for Finding the Best Cloud VM

This work identifies the fragility problem in applying Bayesian Optimization in searching for the best cloud configuration. We propose a low-level augmented Bayesian Optimization method to alleviate the fragility problem. Based on this work, we conclude that it is often insufficient to use general-purpose off-the-shelf methods for configuring cloud instances without augmenting those methods with essential systems knowledge such as CPU utilization, working memory size and I/O wait time.

Chin-Jung Hsu, Vivek Nair, Vincent W. Freeh, Tim Menzies

In ICDCS 2018 (submitted), 2017.

Inside-Out: Reliable Performance Prediction for Distributed Storage Systems in the Cloud

Software-defined storage requires to meet users’ performance requirements. Machine learning techniques are used to create reliable performance models from low-level system metrics collected at runtime. The accurate performance model enables service providers to provision storage resources in a more fine-grained way.

Chin-Jung Hsu, Rajesh K Panta, Moo-Ryong Ra, Vincent W. Freeh

In SRDS (Best Paper Award), 2016.