时间：2007年6月22日上午9：30
地点：曹光彪科技大楼201会议室
Presentation Title: Model-Driven Automation in Utility Computing Environments
Abstract:
This talk is composed of two parts. In Part I, I’ll give an overview of the research activities in my Utility Infrastructure Management Department within the Enterprise Systems and Software Lab, especially in the area of model-driven automation in data centers. In particular, I’ll highlight a number of successful research projects, including policy-based service configuration, capacity management, optimized resource assignment, automated diagnosis, as well as adaptive monitoring. I’ll also describe some case studies where these technologies have been used.
In Part II, I’ll introduce our project ACTS: Automated Control and Tuning of Systems. This work is motivated by the need to enable enterprise IT applications to meet their Service Level Objectives (SLOs) in the presence of system condition changes and short term demand fluctuations, at the same time, to maintain high utilization of data center resources. Prior work in this area has largely relied on ad-hoc techniques or statistical and/or optimization methods. In this work, we propose research aimed at understanding how resource allocation and performance tuning of IT applications can be automated using formal feedback control theory-based approaches. Finally I’ll also talk about our ongoing work and future research directions.
Bio:
Xiaoyun Zhu is a senior research scientist in the Enterprise Systems and Software Lab at Hewlett-Packard Laboratories in Palo Alto, California. Dr. Zhu received her Ph.D. in Electrical Engineering from California Institute of Technology in 2000, specializing in control theory and applications. Her research at HP Labs has focused on automating resource allocation and service level management of IT applications in data centers using mathematical optimization, algorithms and heuristics, statistical analysis and simulation, as well as control theory. Dr. Zhu has published over 25 technical papers and filed over 20 U.S. patent applications. She’s been on the program committee of NOMS’06, ICAC’07, IM’07, DSOM’07, and NOMS’08, and has co-chaired the first International Workshop on Feedback Control Implementation and Design in Computing Systems and Networks (FeBID2006).
地点：曹光彪科技大楼201会议室
Presentation Title: Model-Driven Automation in Utility Computing Environments
Abstract:
This talk is composed of two parts. In Part I, I’ll give an overview of the research activities in my Utility Infrastructure Management Department within the Enterprise Systems and Software Lab, especially in the area of model-driven automation in data centers. In particular, I’ll highlight a number of successful research projects, including policy-based service configuration, capacity management, optimized resource assignment, automated diagnosis, as well as adaptive monitoring. I’ll also describe some case studies where these technologies have been used.
In Part II, I’ll introduce our project ACTS: Automated Control and Tuning of Systems. This work is motivated by the need to enable enterprise IT applications to meet their Service Level Objectives (SLOs) in the presence of system condition changes and short term demand fluctuations, at the same time, to maintain high utilization of data center resources. Prior work in this area has largely relied on ad-hoc techniques or statistical and/or optimization methods. In this work, we propose research aimed at understanding how resource allocation and performance tuning of IT applications can be automated using formal feedback control theory-based approaches. Finally I’ll also talk about our ongoing work and future research directions.
Bio:
Xiaoyun Zhu is a senior research scientist in the Enterprise Systems and Software Lab at Hewlett-Packard Laboratories in Palo Alto, California. Dr. Zhu received her Ph.D. in Electrical Engineering from California Institute of Technology in 2000, specializing in control theory and applications. Her research at HP Labs has focused on automating resource allocation and service level management of IT applications in data centers using mathematical optimization, algorithms and heuristics, statistical analysis and simulation, as well as control theory. Dr. Zhu has published over 25 technical papers and filed over 20 U.S. patent applications. She’s been on the program committee of NOMS’06, ICAC’07, IM’07, DSOM’07, and NOMS’08, and has co-chaired the first International Workshop on Feedback Control Implementation and Design in Computing Systems and Networks (FeBID2006).
