Research Interests
Parallel
and distributed systems, storage systems, real time systems, embedded systems,
and performance evaluation
Projects
Current Work
A large number of existing parallel storage systems consist of hybrid storage components, including solid-state drives (SSD), hard disks (HDD), and tapes. Compared with high-speed storage components (e.g. SSD and HDD), tapes inevitably become an I/O performance bottleneck. Prefetching and caching are commonly employed techniques to boost I/O performance by increasing the data hitting rate of high-end storage components. However, prefetching in the context of hybrid storage systems is technically challenging due to an interesting dilemma: aggressive prefetching schemes can efficiently reduce I/O latency, whereas overaggressive schemes may waste I/O bandwidth by transferring useless data from HDDs to SSDs or from tapes to HDDs. In this research project, we will investigate new data-mining-based multilayer prefetching techniques to improve performance of hybrid storage systems. The goals of this research are to (1) design data-mining algorithms for multilayer prefetching; (2) develop predictive parallel prefetching mechanism for SSD-based storage systems; (3) implement parallel data transfer among SSDs, HDDs, and tapes; (4) develop meta-data management schemes; and (5) implement a simulation framework named FastStor-SIM. The developed toolkit can be used to improve the I/O performance of data centers with hybrid storage systems.
Power,
Security, and Performance Issues in Storage Systems
Power, security, and performance evaluation of storage systems are of critical importance. This work is intended to develop new adaptive strategies that can judiciously select appropriate power states and security services for disk I/O requests while endeavoring to guarantee various performance requirements ( e.g., desired response times). In what follows, the novel approaches to secure storage systems are described.
Previous Work
An Adaptive
Write Strategy for Secure Local Disk Systems
.4 Protect stored data from being tampered or disclosed. Although an increasing
number of secure storage systems have been developed, there is no way to
dynamically choose security services to meet disk requests' flexible security
requirements. Furthermore, existing security techniques for disk systems are not
suitable to guarantee desired response times of disk requests. We remedy this
situation by proposing an adaptive strategy (referred to as AWARDS) that can
judiciously select the most appropriate security service for each write request
while endeavoring to guarantee the desired response times of all disk requests.
Experimental results show that AWARDS significantly improves security and
overall performance over an existing scheme by up to 325.0% and 358.9% (with
averages of 199.5% and 213%).
Quality of
Security Control in Parallel and Distributed Disk Systems
Building networked and data intensive applications are highly scalable and can
alleviate the problem of disk I/O bottleneck. Although a number of parallel and
distributed disk systems have been developed, the systems lack a means to
optimize quality of security for dynamically changing networked environments. We
remedy this situation by proposing an adaptive quality of security control
scheme for parallel and distributed disk systems that makes it possible for disk
systems to adapt to changing security requirements and workload conditions. Our
approach is carried out in three phases: dynamic data partitioning, response
time estimation, and adaptive security quality control. Hence, our scheme is
conducive to adaptively and expeditiously determining security schemes for disk
requests in a way to improve security of parallel and distributed disk systems
while making an effort to guarantee desired response times of the requests.