Garth Gibson. "Redundant Disk Arrays: Reliable, Parallel Secondary Storage". PhD thesis, University of California at Berkeley, 1990. Report UCB/CSD 91/613. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Proving Correctness of a Controller Algorithm for the RAID.. - Vaziri, Lynch (1997) (1 citation) (Correct)
....disk array may decrease dramatically [Gibson93] We consider independent catastrophic disk failures, i.e. disk failures in which all data stored on the disk becomes inaccessible and the disk cannot be written any further. RAID systems are designed to be fault tolerant by storing redundant data [Gibson90] on extra disks and to tolerate 1 or 2 disk failures. The redundancy can be an identical copy of each disk, also known as disk mirroring [Bitton88, Gray90] In this case, if the disk containing one copy fails, the controller can use the other copy which is on a separate disk. Having two copies of ....
Garth Gibson. "Redundant Disk Arrays: Reliable, Parallel Secondary Storage". PhD thesis, University of California at Berkeley, 1990. Report UCB/CSD 91/613.
Reducing the Cost of System Administration of a Disk Storage.. - Asami (2000) (18 citations) (Correct)
....classes of disk storage systems that are related to my research; hardware RAID, high availability server systems and network attached storage systems. Hardware RAID RAID (Redundant Array of Inexpensive Disks) is a method of building large disk storage systems from smaller disks [CGP 88, Gib92] There are six levels of RAIDs, 0 through 5, differing in redundancy schemes. Most widely used are RAID 0, which is simple striping with no redundancy, RAID 1, which is mirroring, and RAID 5, which is block interleaved parity without a dedicated parity disk. There are many companies EMC, Sony, ....
Garth Gibson. Redundant Disk Arrays: Reliable Parallel Secondary Storage. The MIT Press, 1992.
The Design of Large-Scale, Do-It-Yourself RAIDs - Asami, Talagala, Anderson.. (1995) (Correct)
....missing data onto the spare rather than waiting for the end user to replace the failed disk. Hot spares thus reduce the time the system is vulnerable to another failure. These standby disks need not be empty; spreading the equivalent space among all disks means the extra disks can help performance [17]. RAIDs have two drawbacks. If the system does not simultaneously write all G 1 data blocks at once, then each write turns into four disk accesses: read the old data and old parity to calculate new parity and then write the new data and new parity. Several techniques exist for batching writes in ....
....next section evaluates the reliability of that configuration. We assume the load is balanced across the disks, and that the 16 hot spares really consist of space being reserved across 1120 disks so that the system can gain the benefit of both hot spares and the performance advantage of more disks [17]. Similar to Figure 5, Figure 8 shows the costs for the I O components for the other string interfaces. We chose current prices except for SSA and FC AL because it would be unfair to SSA and FC AL given their youth. For example, a FC AL adapter costs 4000 in 1995. Hence we optimistically use ....
[Article contains additional citation context not shown here]
Garth Alan Gibson. Redundant Disk Arrays: Reliable Parallel Secondary Storage. Ph.D. thesis, University of California at Berkeley, April 1990.
Performance of a Disk Array Prototype - Chervenak, Katz (1991) (13 citations) (Correct)
No context found.
Garth A. Gibson, "Redundant Disk Arrays: Reliable, Parallel Secondary Storage", U.C. Berkeley Technical Report UCB/CSD 90/613, Berkeley, CA, March, 1991, Ph.D. Dissertation.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC