RAID 60 Storage Reliability and Cost Optimization Using Weibull Distribution and Cuckoo Search Algorithm with Lévy Flights

RAID 60 Storage Reliability and Cost Optimization Using Weibull Distribution and Cuckoo Search Algorithm with Lévy Flights

Mhamed Zineddine (INSA-EUROMED, Fes, Morocco) and Adel Akaaboune (School of Business Administration, ALHOSN University, Abu Dhabi, United Arab Emirates)
Copyright: © 2018 |Pages: 27
DOI: 10.4018/IJAMC.2018070102

Abstract

This article describes how in this digital age, storage devices have been permitting the design of systems with the capacity to handle massive amount of diverse types of data. Real-time storage devices have been using complex architectures to accommodate the exponential increase of generated content. Hard drives have been at the core of most of storage systems. Estimating the reliability and the time to failure of such devices will be of great value. Weibull distribution has been widely used to assess the reliability of many systems. In this article, the three parameter Weibull distribution and Cuckoo search algorithm are combined to optimize the reliability and the maintenance cost of a RIAD 60 storage system. The numerical results show that using the proposed approach in this article, a RAID 60 system could reach its optimal reliable state by swapping less reliable drives by others more (used or new) reliable, whichever, is optimal according to the sought-after reliability threshold.
Article Preview
Top

2. Background

2.1. Storage Devices

Cloud computing and other extensive storage demanding applications have been placing ever-greater demands on storage systems. Surveillance systems placed in private and public places are generating massive video content that need to be stored and processed real-time or later as situations warrant (Sahu et al., 2011). Data mining for patterns, face recognition algorithms, and other data extensive applications have been requiring higher processing power and fast access to data. For instance, face recognition applications have to process captured video-feed real-time in order to identify criminals walking in train stations or other public areas (Zhang et al., 2015; Zhang et al., 2014).

Redundant Array of Inexpensive Disks (RAID) has been a dominant technology for performance improvement and reliability of storage systems. A RAID system is an arrangement of two or more independent drives configured to work in parallel (Chen et al., 1994). Standard hard disks are the most used; however, the use of SSD technology has recently been flourishing. Different RAID configurations are optimized for different situations. RAID implementations include striping (RAID 0), mirroring (RAID 1), striping with parity (RAID 5), and striping with double parity (RAID 6). Moreover, more advanced configurations called Nested (Hybrid) RAID (Vijayan, et al. 1995) or top arrays have been developed to accommodate more size, performance, and reliability. For instance, storage systems might combine mirroring and striping, originally denoted as RAID 1+0, which yields RAID 10 when the + is omitted. Similarly, combining the distributed parity of RAID 5 and the striping of RAID 0 yields RAID 50 (Layton, 2011).

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 11: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 4 Issues (2016)
Volume 6: 4 Issues (2015)
Volume 5: 4 Issues (2014)
Volume 4: 4 Issues (2013)
Volume 3: 4 Issues (2012)
Volume 2: 4 Issues (2011)
Volume 1: 4 Issues (2010)
View Complete Journal Contents Listing