M. Holland, G. Gibson, and D. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. In FTCS-23, France, 1993. Home/Search Document Not in Database Summary Related Articles Check |
This paper is cited in the following contexts:
Analysis of Methods for Scheduling Low Priority Disk Drive.. - Bachmat, Schindler (2002) (3 citations) (Correct)
....when the system is always busy or when the background task must maintain a certain rate of forward progress) the background requests have negative impact on the performance. There are numerous examples of solutions that analytically express the impact of background tasks on foreground activity [4, 8, 9, 22]. All these solutions focus on tasks that fall into a class of algorithms that proceed sequentially from the beginning of the disk. We call these algorithms ordered. In this paper, we analyze a class of greedy algorithms that minimize the negative impact on response times of foreground requests ....
....such as scrubbing, data backup, and rebuild, or to tasks that access only a portion of a disk such as disk reorganization and virus detection. Reducing the completion time and minimizing the performance impact on foreground activity is particularly valuable for data rebuild in RAID configurations [4, 10]. The observed 19.2 faster rebuild time with our greedy algorithm shortens the window of opportunity for data loss due to another disk failure when the system is in degraded mode (i.e. rebuilding data after a disk failure) In addition, our greedy algorithm offers 10 shorter foreground request ....
[Article contains additional citation context not shown here]
Mark Holland, Garth A. Gibson, and Daniel P. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. 23rd International Symposium on Fault-Tolerant Compter Systems (Toulouse, France, 22--24 June 1993.
Analysis of Methods for Scheduling Low Priority Disk Drive.. - Bachmat, Schindler (2002) (3 citations) (Correct)
....when the system is always busy or when the background task must maintain a certain rate of forward progress) the background requests have negative impact on the performance. There are numerous examples of solutions that analytically express the impact of background tasks on foreground activity [4, 8, 9, 22]. All these solutions focus on tasks that fall into a class of algorithms that proceed sequentially from the beginning of the disk. We call these algorithms ordered. In this paper, we analyze a class of greedy algorithms that minimize the negative impact on response times of foreground requests ....
....such as scrubbing, data backup, and rebuild, or to tasks that access only a portion of a disk such as disk reorganization and virus detection. Reducing the completion time and minimizing the performance impact on foreground activity is particularly valuable for data rebuild in RAID configurations [4, 10]. The observed 19.2 faster rebuild time with our greedy algorithm shortens the window of opportunity for data loss due to another disk failure when the system is in degraded mode (i.e. rebuilding data after a disk failure) In addition, our greedy algorithm offers 10 shorter foreground request ....
[Article contains additional citation context not shown here]
Mark Holland, Garth A. Gibson, and Daniel P. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. 23rd International Symposium on Fault-Tolerant Compter Systems (Toulouse, France, 22--24 June 1993.
Architectures and Algorithms for Scalable Wide-area Information.. - Tewari (1998) (3 citations) (Correct)
....single cycle, but plays it out over n cycles, roughly halving the memory requirements. Their non clustered scheme has the entire stripe group read in a cycle, only under failure conditions. Similarly, there are numerous other research efforts on RAID, mirrored data layout, and declustering schemes [16, 21, 93, 60, 61], but these do not focus on clusters with heterogeneous data types. Previous work that considers high availability in a clustered environment that we know of is by Frey [44] and the work of Haskin [59] Frey et al. describe a round robin layout across nodes and disks, but they do not consider ....
M. Holland, G. A. Gibson, and D. P. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. Proceedings of FTCS, 1993.
Fault Tolerance Issues in Data Declustering for Parallel.. - Golubchik, Muntz (1994) (Correct)
.... the following aspects of the disk subsystem must be examined: a) performance under normal operation (e.g. 7, 31] b) mean time to data loss (or system failure) e.g. 14] and c) performance of the disk subsystem under failure, i.e. when one or more disks are inoperable or inaccessible (e.g. [35, 30, 17, 19, 18]) We should keep in mind that failures are expected to occur relatively infrequently, so most of the time a system is in a fully operational mode. Thus, it is important to provide reliability techniques that do not (significantly) hinder the system s performance during normal operation. Since ....
M. Holland, G. A. Gibson, and D. P. Siewiorek. Fast, On-Line Failure Recovery in Redundant Disk Arrays. In 23rd Annual International Symposium on Fault-Tolerant Computing, 1993.
Improving Storage System Availability with D-GRAID - Muthian Sivathanu Vijayan (2004) (1 citation) (Correct)
No context found.
M. Holland, G. Gibson, and D. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. In FTCS-23, France, 1993.
Improving Storage System Availability with D-GRAID - Sivathanu, Prabhakaran.. (2004) (1 citation) (Correct)
No context found.
M. Holland, G. Gibson, and D. Siewiorek. Fast, on-line failure recovery in redundant disk arrays. In FTCS-23, France, 1993.
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