Founded in 1966

 

David Essary (Pitt/CS)

PhD Proposal Defense

Friday, October 19, 2007
10:30pm - SENSQ 6106 - Eli Lilly

Abstract

Data retrieval latencies are considered a major bottleneck, and with growing volumes of data and increased storage needs it is only growing in significance. Data storage infrastructure is therefore a growing consumer of energy at data center scales, while the individual disk is already a significant concern for mobile computing (accounting for almost a third of a mobile system's energy demands).
While improving responsiveness of storage subsystems and reducing latencies in data retrieval is often considered contradictory with efforts to reduce disk energy consumption, we demonstrate that predictive data grouping has the potential to simultaneously work towards both these goals. Predictive data grouping has advantages in its applicability compared to both prior approaches to reducing latencies and prior approaches to reducing energy usage. For latencies, grouping can be performed opportunistically, and thereby avoids the serious performance penalties that can be incurred with prior applications of access prediction (such as predictive prefetching of data). For energy, we show how predictive grouping can even save energy use for an individual disk that is never idle.

Our proposed work builds upon our initial success with static grouping utilizing adaptive grouping with increased space efficiency as well as more space-efficient access tracking strategies.  These tracking strategies also show promise in adapting to changing workloads while maintaining robustness against noise, or seemingly random accesses.  Our proposed work also includes development of a single device prototype as well as verification on a multi-device system.

Dissertation Adviser

Prof. Ahmed Amer, Department of Computer Science

Committee Members

Prof. Panos K. Chrysanthis, Department of Computer Science
Prof. Kirk Pruhs, Department of Computer Science
Prof. Patchrawat Uthaisombut, Department of Computer Science
Dr. Alex Jones, Electrical Engineering

You are using an older browser that does not support current Web standards. Although this site is viewable in all browsers, it will look much better in a browser that supports Web standards.