Cisco Systems MDS 9000 Manual De Usuario

Cisco Systems, Inc.

All contents are Copyright © 1992–2002 Cisco Systems, Inc. All rights reserved. Important Notices and Privacy Statement.

Page 6 of 11

A storage network infrastructure provides multiple paths to storage subsystems, including disks and tapes, for both

high availability and scalability. Customers can implement improved disaster recovery solutions, especially within an

open systems environment. In legacy implementations, disaster recovery at a remote site is typically implemented by

server-to-server communications over a LAN. In a networked model, data can also be mirrored between two storage

subsystems freeing expensive server and LAN resources. Migrating backup traffic onto a storage network limits

potential outages within each domain thereby protecting backups from any failure in LAN data traffic flow and

vice versa.

Storage consolidation allows multiple servers to share the same storage devices and reduces the number of tape

libraries required in the data center. Consolidation also makes it easier to reassign unused capacity amongst all

servers to increase utilization and efficiency rates. Customers can gain additional savings by implementing an

enterprise backup/recovery solution to reduce management and maintenance costs associated with server-attached

tape drives. The cost of managing each individual component can be worsened by a high potential for human error

resulting from manual processes. Mechanisms such as remote tape vaulting that deploy robots to eliminate manual

transport of tapes between sites help reduce these errors. This also enhances reliability by eliminating damage caused

by handling, potential for loss, and availability since the data that needs to be restored is never in transit. Storage

networks enable pooling of backup resources that each server can draw upon without the limitations caused by a

failure of an individual backup device. Lower total cost of ownership (TCO) is also delivered via enhanced scalability,

availability, performance, and manageability of shared backup resources.

Businesses are now open for longer hours and support a global community of customers which drives requirements

for round-the-clock operation. The two backup options are available including hot and cold backup. Cold backup

relates to the scenario where application data is tied up for the duration of the backup. However, a hot backup relates

to the scenario where a system is performing this operation while applications are updating data.

Hot backup technologies such as copy-on-write and split-mirror snapshot use a mirrored image of the original data

created at a particular instant to backup online without affecting the application. Both copy-on-write and

split-mirror options copy data blocks to unused storage to create a point-in-time copy and are supported by most

databases. The copy-on-write and split-mirror options manage the mapping process of physical data blocks and their

correlation to a file system or database. The storage subsystem vendors that support these options include EMC

Timefinder using the Business Continuance Volume (BCV), HDS ShadowImage, and FlashCopy in the IBM

Enterprise Storage Server (ESS) or the Modular Storage server (MSS).

Backup implementations strive to reduce the time required for backups to minimize disruption to user traffic

(especially for cold backups) and increase performance. Hot backups utilize mirroring, whether it be local or

remote, to ensure continuous availability for applications. A detailed discussion of remote mirroring technology

is provided below.