The constant evolution of new varieties of computing systems - cloud computing, mobile devices, and Internet of Things, to name a few - have necessitated a growing need for highly reliable, available, secure, and high-performing storage systems. While CPU performance has typically scaled with Moore's Law, data storage is much less consistent in how quickly performance increases over time. One method of improving storage performance is through the use of special storage architectures. Such architectures often include redundant arrays of independent disks (RAID). RAID provides a meaningful way to increase storage performance on a variety of levels, some higher-performing than others. The fastest performing arrays, however, come at the expense of data reliability, preventing their use in many applications. In this thesis, a storage architecture is presented which utilizes the highest performing RAID architecture (RAID 0) combined with highly available cloud storage in order to maximize storage performance and reliability, while providing an additional benefit of security that is not currently provided by similar storage architectures. Various levels of RAID and the current field of RAID and RAID-like technologies are outlined. Furthermore, the benefits of combining local and cloud storage in RAID configurations, including the ability to recover data in the event of a simultaneous failure of all local disks within an array, are discussed. Reliability calculations are performed, and a 3x Mean Time to Failure (MTTF) improvement over existing systems is shown. The MTTF, privacy, read/write throughput, and usable storage percentage improvements of local+cloud-RAID 4 and local+cloud-RAID-DP are discussed. Thereafter, a software implementation of local+cloud-RAID 4 is presented and discussed, allowing for notable benefits over existing architectures in terms of reliability and privacy while maintaining performance. Finally, a novel hardware-based local+cloud-RAID controller architecture is described and discussed.
College and Department
Ira A. Fulton College of Engineering and Technology; Electrical and Computer Engineering
BYU ScholarsArchive Citation
Hansen, Christopher Glenn, "A Secure, Reliable and Performance-Enhancing Storage Architecture Integrating Local and Cloud-Based Storage" (2016). All Theses and Dissertations. 6470.
RAID, storage, security, privacy, reliability, architecture