Browse KB by category:
Go to KB #:
What is disk staging and how is it implemented?
Disk Staging (Disk-to-Disk-to-Tape)
The Challenge - To minimize the interruption of normal system activity while, at the same time, providing mechanisms to protect and reliably restore data lost when either the primary server system or a client system fails.
The Solution - BRU Server Disk Staging in concert with core BRU™ functionality reduces the perceived backup window at the client level, reduces the backup verification window at the server level, and provides the high speed recovery of files from disk, rather than tape.
BRU Server Disk Staging - The concept of disk staging is the writing of backup data to hard disk first rather than directly to tape, traditionally a slower technology. Then, at the system administrator's discretion, the data contained in the disk staging area is committed to tape in a background process.
BRU Server Disk Staging uses a centralized disk farm attached to the server system running BRU Server. The Disk Staging area can be configured as all, or part of an attached Xserve RAID subsystem, or on a simple external Firewire disk. Because disk access is asynchronous, BRU Server's Staging allows multiple client systems to send their backup data streams simultaneously to minimize the effect of backup window downtime. Performance is limited only by the network's bandwidth.
BRU Server Disk Staging can be implemented on any filesystem that is available to the physical server system - locally attached, Fibre-Channel SAN, or Network attached (depending on network bandwidth). BRU Server incurs very little overhead and can effectively be installed on the primary network server as shown in Figure 1 when the server is not task intensive. Backup data stream paths from client groups A, B and C can be directed to the Staging area.
When the primary server is task intensive, it is suggested BRU Server and its associate Disk Staging area be implemented as a separate backup system server attached on the network as shown in Figure 2. The separate BRU Server backup system can be implemented on a G4, G5, or an Xserve system. Whether or not the primary server is task intensive, implementing a separate backup system server is recommended because it removes the data protection function from the primary server's duties and provides the highest level of performance.
The benefits of implementing BRU Server Disk Staging include:
A Caveat - The ultimate goal of backing up data is to protect it from irreplaceable loss. While writing to a disk farm staging area provides significant benefit to minimize the intrusion of a backup process, the ultimate commitment of data to tape provides the long-term assurance of data availability and recoverability. A good tape backup strategy that includes off-site tape storage in concert with disk staging delivers the highest backup system performance, reliability and data availability.
|Others in this Category|
|MegaBytes versus Megabits|
|Understanding Unix/Linux Timestamp's|
|Disable/Enable Hardware Compression When Using BRU|
|What are TapeAlert messages and the T10 error codes?|
|Tape Drive Performance Figures|
|» More Articles|