Current Research: Cluster resource management
My PhD thesis work involves resource management in multi-tenant shared clusters. The aim is to improve cluster resource utilization while also improving overall performance and quality of service for users. My on-going focuses on adaptive application scalability and resource-aware scheduling. Most recently, our paper on preemptive scheduling in shared clusters was accepted by Middleware.
My research in the Elba project involves studying application performance on different clouds and virtualization platforms. Previously, I studied multi-bottlenecks in n-tier applications using machine learning. The Elba project's goal is automation of large-scale application system management from deployment to production as well as system monitoring, evaluating performance, and diagnosing problems.
In the Ques project, I explored various ways for databases to self-heal by defining all classes of failures and developing mechnisms to remedy each class of failure. In addition, I researched different rebalancing schemes in Hadoop Distributed File System and added an implementation for a "hot" file rebalancing mechanism to the existing HDFS rebalancer.
My research in the SyNRG group dealt mainly with mobile applications--particularly localization. I was part of the Micro-Blog project which uses mobile phones as a virtual information telescope that allowed users to zoom into anywhere in the world and observe events in that area. Also, I explored ways for users to localize their logical location using acceleromters and compasses found on many of today's commercial smart phones.
My complete publication list can be found on Google Scholar.
Jack Li, Yuan Chen, Daniel Gmach, Dejan Milojicic, Calton Pu. "Enabling Elastic Stream Processing in Shared Clusters". IEEE Cloud 2016. To Appear
Jack Li, Yuan Chen, Vanish Talwar, Calton Pu, Dejan Milojicic. "Improving Preemptive Scheduling with Application-Transparent Checkpointing in Shared Clusters". ACM Middleware 2015. [Slides]
Qingyang Wang, Yasuhiko Kanemasa, Jack Li, Chien-An Lai, Chien-An Cho, Yuji Nomura, Calton Pu. "Lightning in the Cloud: A Study of Very Short Bottlenecks on n-Tier Web Application Performance". USENIX TRIOS 2014.
Jack Li, Qingyang Wang, Chien-An Lai, Junhee Park, Daisuke Yokoyama, Calton Pu. "The Impact of Software Resource Allocation on Consolidated n-Tier Applications". CLOUD 2014.
Qingyang Wang, Yasuhiko Kanemasa, Jack Li, Chien-An Lai, Masazumi Matsubara, Calton Pu. "Impact of DVFS on n-Tier Application Performance". ACM TRIOS 2013.
Qingyang Wang, Yasuhiko Kanemasa, Jack Li, Deepal Jayasinghe, Toshihiro Shimizu, Masazumi Matsubara, Motoyuki Kawaba, Calton Pu. "Detecting Transient Bottlenecks in n-Tier Applica-tions through Fine-Grained Analysis". IEEE ICDCS 2013.
Jack Li, Qingyang Wang, Deepal Jayasinghe, Junhee Park, Tao Zhu, Calton Pu. "Performance Overhead Among Three Hypervisors: An Experimental Study using Hadoop Benchmarks". IEEE BigData Congress 2013.
Deepal Jayasinghe, Galen Swint, Simon Malkowski, Jack Li, Qingyang Wang, Junhee Park and Calton Pu. "Expertus: A Generator Approach to Automate Performance Testing in IaaS Clouds". IEEE CLOUD 2012.
Jack Li, Qingyang Wang, Yasuhiko Kanemasa, Deepal Jayasinghe, Simon Malkowski, Pengcheng Xiong, Motoyuki Kawaba and Calton Pu. "Profit-Based Experimental Analysis of IaaS Cloud Performance: Impact of Software Resource Allocation". IEEE SCC 2012.
Jiaxing Zhang, Hucheng Zhou, Rishan Chen, Zhenyu Guo, Haoxiang Lin, Jack Li, Wei Lin, Jingren Zhou, and Lidong Zhou. "Optimizing Data Shuffling in Data-Parallel Computation by Understanding User-Defined Funtions". USENIX NSDI 2012.
Deepal Jayasinghe, Simon Malkowski, Qingyang Wang, Jack Li, Pengcheng Xiong, and Calton Pu. "Variations in Performance and Scalability when Migrating n-Tier Applications to Different Clouds". IEEE CLOUD 2011.
Shravan Gaonkar, Jack Li, Romit Roy Choudhury, Landon Cox, Al Schmidt. "Micro-Blog: Sharing and Querying Content through Mobile Phones and Social Participation". ACM MobiSys 2008.