Research interests:
Storage systems, Real-time and Embedded Systems, Fault tolerance, Reliability, Security and Energy Efficiency
Overview
My research mainly focuses on the energy conservation and reliability issues of large-scale storage systems. More specifically, I have been investigating energy-efficient strategies and reliability-aware framework to improve power efficiency and reliability of large-scale parallel disk systems
Reliability Aware Energy efficient Storage systems
Numerous energy saving techniques have been developed to aggressively reduce energy dissipation in parallel disks. However, many existing energy conservation schemes have substantial adverse impacts on the reliability of disks. To remedy this deficiency, we address the problem of making tradeoffs between energy efficiency and reliability in parallel disk systems with data mirroring. Among several factors affecting disk reliability, the most two important factors, disk utilization and disk age are the focus of this study. We build a mathematical reliability model to quantify the impacts of disk age and utilization on failure probabilities of mirrored disk systems. In light of the reliability model, we propose a novel concept of safe utilization zone, within which energy dissipation in disks can be reduced without degrading reliability. We developed an approach to improving both reliability and energy efficiency of disk systems through disk mirroring and utilization control that enforces disk drives to be operated in safe utilization zones. This is the first utilization-based control scheme that seamlessly integrates reliability with energy saving techniques in the context of fault-tolerant systems.
Security-Aware Fault-Tolerant Real-Time Systems
Real time applications such as military aircraft flight control systems and online banking systems are critical with respect to security and reliability. Although many conventional fault-tolerant or security approaches were investigated and applied to real-time systems, most existing schemes only addressed either security demands ignoring the fault-tolerant requirements or vice versa. To bridge this technology gap in real-time systems, we proposed a way to integrate security services and reliability for security-aware and high-reliable real-time systems.
Acknowledgement
The work reported in this paper was supported by the US National Science Foundation under Grants No. CCF-0742187, No. CNS-0757778, No. CNS-0831502, No. OCI-0753305, and No. DUE-0621307, and Auburn University under a startup grant