Abstract-We present a novel method, that we call EVEN ODD, for tolerating up to two disk failures in RAID architectures. EVEN ODD employs the addition of only two redundant disks and consists of simple exclusive-OR computations. This redundant storage is optimal, in the sense that two failed disks cannot be retrieved with less than two redundant disks. A major advantage of EVENODD is that it only requires parity hardware, which is typically present in standard RAID-S controllers. Hence, EVENODD can be implemented on standard RAID-S controllers without any hardware changes. The most commonly used scheme that employes optimal redundant storage (i.e., two extra disks) is based on Reed-Solomon (RS) error-correcting codes. This scheme requires computation over finite fields and results in a more complex implementation. For example, we show that the complexity of implementing EVENODD in a disk array with 15 disks is about 50% of the one required when using the RS scheme.The new scheme is not limited to RAID architectures: it can be used in any system requiring large symbols and relatively short codes, for instance, in multitrack magnetic recording. To this end, we also present a decoding algorithm for one column (track) in error.

No abstract

We present a novel method, that we call EVEN-

Today's data storage systems are increasingly adopting low-cost disk drives that have higher capacity but lower reliability, leading to more frequent rebuilds and to a higher risk of unrecoverable media errors. We propose an efficient intradisk redundancy scheme to enhance the reliability of RAID systems. This scheme introduces an additional level of redundancy inside each disk, on top of the RAID redundancy across multiple disks. The RAID parity provides protection against disk failures, whereas the proposed scheme aims to protect against media-related unrecoverable errors. In particular, we consider an intradisk redundancy architecture that is based on an interleaved parity-check coding scheme, which incurs only negligible I/O performance degradation. A comparison between this coding scheme and schemes based on traditional Reed--Solomon codes and single-parity-check codes is conducted by analytical means. A new model is developed to capture the effect of correlated unrecoverable sector errors. The probability of an unrecoverable failure associated with these schemes is derived for the new correlated model, as well as for the simpler independent error model. We also derive closed-form expressions for the mean time to data loss of RAID-5 and RAID-6 systems in the presence of unrecoverable errors and disk failures. We then combine these results to characterize the reliability of RAID systems that incorporate the intradisk redundancy scheme. Our results show that in the practical case of correlated errors, the interleaved parity-check scheme provides the same reliability as the optimum, albeit more complex, Reed--Solomon coding scheme. Finally, the I/O and throughput performances are evaluated by means of analysis and event-driven simulation.

No abstract

No abstract

RISC vs. CISC wars raged in the 1980s when chip area and processor design complexity were the primary constraints and desktops and servers exclusively dominated the computing landscape. Today, energy and power are the primary design constraints and the computing landscape is significantly different: growth in tablets and smartphones running ARM (a RISC ISA) is surpassing that of desktops and laptops running x86 (a CISC ISA). Further, the traditionally low-power ARM ISA is entering the high-performance server market, while the traditionally high-performance x86 ISA is entering the mobile low-power device market. Thus, the question of whether ISA plays an intrinsic role in performance or energy efficiency is becoming important, and we seek to answer this question through a detailed measurement based study on real hardware running real applications. We analyze measurements on the ARM Cortex-A8 and Cortex-A9 and Intel Atom and Sandybridge i7 microprocessors over workloads spanning mobile, desktop, and server computing. Our methodical investigation demonstrates the role of ISA in modern microprocessors' performance and energy efficiency. We find that ARM and x86 processors are simply engineering design points optimized for different levels of performance, and there is nothing fundamentally more energy efficient in one ISA class or the other. The ISA being RISC or CISC seems irrelevant.

As the amount of data being stored in the open systems environment continues to grow, new paradigms for the attachment and management of data and the underlying storage of the data are emerging. One of the emerging technologies in this area is the storage area network (SAN). Using a SAN to connect large amounts of storage to large numbers of computers gives us the potential for new approaches to accessing, sharing, and managing our data and storage. However, existing operating systems and file systems are not built to exploit these new capabilities. IBM Storage Tank TM is a SANbased distributed file system and storage management solution that enables many of the promises of SANs, including shared heterogeneous file access, centralized management, and enterprise-wide scalability. In addition, Storage Tank borrows policybased storage and data management concepts from mainframe computers and makes them available in the open systems environment. This paper explores the goals of the Storage Tank project, the architecture used to achieve these goals, and the current and future plans for the technology. IBM Storage Tank* (ST) is a multiplatform, scalable file system and storage management solution that works with storage area networks (SANs). By means of SANs, thousands of computers can connect to and share a large number of storage devices that range from simple disks to large, high-performance, highfunction storage systems. The current state of SAN technology limits its use to machine room environments; therefore, ST is also currently limited in the same way. As SANs evolve beyond machine room environments, so will ST.ST goes beyond cluster file systems, such as the IBM General Parallel File System (GPFS), 1 that allow a cluster of homogenous (single operating system) computers to share data by allowing thousands of heterogeneous computers, some subset of which may be clustered, to share data. ST can provide an effective solution for customers with as little as tens of computers and a terabyte of data, and can scale up to support customers with thousands of computers, petabytes of data, and billions of files.In addition to sharing data, ST also centralizes storage management functions such as backup, restore, and file allocation. This centralization replaces the labor-intensive, computer-by-computer storage management that is currently in practice. ST further simplifies storage management by supporting policy-based storage management. An administrator specifies policies for how backup, restore, allocation, and so on are to be performed, and the ST system enforces these policies without human intervention. As a result, ST is an important step in the direction toward autonomic computing. 2 250This paper describes the Storage Tank architecture and design. Whenever we describe the current status of ST in this paper, we are referring to the status of the ST prototype built at the IBM Almaden Research Center. MotivationCustomers face many issues today as they build or grow their storage infrastructures. Although the cost of...