1 Investigating the Reliability in Three RAID Storage Models and Effect of Ordering Replicas on Disks Leila Namvari-Tazehkand a , Saeid Pashazadeh b a,b Faculty of Electrical and Computer Engineering, University of Tabriz, Tabriz, East Azerbaijan, Iran Pashazadeh@tabrizu.ac.ir Abstract One of the most important parts of cloud computing is storage devices, and Redundant Array of Independent Disks (RAID) systems are well known and frequently used storage devices. With the increasing production of data in cloud environments, we need high-reliable storage, given the importance of data. RAID system's reliability analysis is of particular significance in the area of cloud storage. Generally, data redundancy is used to create fault tolerance and increases the reliability of storage. This study has considered three examples of the simple RAID storage models and analyzed their reliability. All of which have a replication factor of two and have the same number of disks and reliabilities. The only difference is the model that they used for generating the redundancy. To compare these three models' reliability, we examined the degree of fault tolerance (FT) and calculated the models' reliability using the reliability block diagram (RBD). In this paper, the effect of redundancy's type and the blocks' arrangement on the system's reliability was investigated. Keywords: RAID, Reliability, Fault Tolerance, Redundancy, Parity Code, Reliability Block Diagram 1. Introduction Many companies have provided cloud storage platforms such as Dropbox, Google Drive, and Amazon EC2, which allow users to enjoy on-demand, anytime, anywhere services from a shared pool of configurable resources in the cloud [1]. In cloud environments, RAID systems are used to store data. The RAID system is one of the most important hardware tools for storing data even at national and global levels. Today, with the expanding use of cloud computing and data storage, the importance of storage systems is increasingly being considered. In a system with thousands of hard disks, disk failure usually occurs, but multi-disk failures are also common. To avoid losing data due to the failure of the disks, there must be a way to think. Therefore, data reliability is one of the main concerns in RAID systems [2]. To have a high level of reliability and fault tolerance in data storages the redundancy method is used. The replica block and parity code block 1 two major styles are redundancy which is used to recover data from failed disks [2,3]. Specialists in the field of data storage software RAID models have introduced a lot which is in each of them for redundancy of information used the replica block method or parity code block method or a combination of the two. Each of the redundancy methods, have their own advantages and disadvantages. In [4] once the data itself and one-time replica block is saved for maintenance of fault tolerance. In [5] in addition to the data itself saved, double replica block and one-time parity code block are saved. In [6,7] in addition to the data itself saved, double replica block is saved. 2. Materials and methods Many studies have been conducted on the redundancy and reliability in RAID systems [1-10]. Table 1 shows the advantages and disadvantages of the two redundancy methods. 1 Parity code is a simple way to add redundancy block by using Xor of two or more blocks.