Almost everyone understands that it is important to back up a computer system. This critical practice ensures the ability to recover from any number of problems, including catastrophic events such as fire, flood, or earthquake.
In day-to-day computer system management, the requirement to restore from a backup will most likely be to restore a strategic file that has become corrupted or accidentally deleted. Less frequently, an entire drive is lost and must be restored as soon as possible. This typically creates a big problem when the lost drive is also the operating system (OS) drive.
Normal file-by-file backups work well for restoring files, except when the failure occurs on a Windows OS disk. Here are a few common failures that can completely cripple a Windows server or workstation: registry file corruption, installing new device drivers, deleting strategic files, and a failed OS hard drive. Any of these problems will render a system unbootable. If the system is unbootable, a regular file-by-file backup cannot be used until a temporary OS has been installed; partitions set; device drivers installed; and when used, 3rd party backup software installed. Only then can a file-based backup recover files to the failed machine.
A file-based OS recovery can take anywhere from two hours to two days depending on the degree of difficulties encountered. Because of these limitations, other methods of backup have been developed to better recover a failed OS. The fastest and easiest to use method is based on image technology.
With image technology, the backup process reads a hard drive on a low-level, bit-by-bit basis. With a built-in locked file backup agent, this process doesn't care what's on the hard drive or even what the hard drive is doing at time of backup. Original image technology didn't see partitions or file system structures and didn't even know when it was backing up data or empty space. Today's image technology backs up on the partition level, excludes work space and files that can't be restored, processes only actively marked in-use disk clusters, uses software compression, and allows media spanning. Even more useful is software that allows (from a recovery perspective) backing up and restoring from network UNC paths, FTP devices, and, for large IBM users, Tivoli Storage Manager (TSM). Being able to back up and recover from network storage enables disaster recovery protection for machines that do not have locally attached tape drives.
While backing up an image is important, even more critical is how to restore that image to a failed system. Originally this was performed by booting from DOS. Today's image technology uses a slick 32-bit Windows-based universal boot CD that requires no setup as did DOS. A universal CD can boot, load all required device drivers, make a network connection, and recover an OS partition in minutes. Recovery can be from local or remote tape, local disk, UNC path, FTP or TSM device. With server downtime costs ranging from hundreds to thousands of dollars a minute, speeding up the recovery of a failed machine by even a few minutes can pay big dividends. On a fast system, users can expect to restore image backups at speeds from 300-400 MB per minute and ultra-fast gigabit networks can see 600-800 MB per minute speeds (second generation LTO drives also restore at this speed when tape is available).
An example of image backup technology is UBDR Pro from UltraBac Software of Bellevue, Washington. UltraBac allows the combined use or standalone operation of file-by-file and image based technology for backup and disaster recovery of Microsoft Windows® servers and workstations (NT/XP/Win2K/Win2003).
UltraBac can integrate both types of backups to the same media or can perform file and image based backups independently to any type of tape or disk media. After installing UltraBac, an administrator creates backup groups and sets to schedule "lights out" backup processing, set parameters, and determine what files or disk partitions to include. Up to 32 concurrent backup sessions can be launched simultaneously. Image backups can also allow restoring individual files provided this option was enabled during backup.
Implementing image restore technology only requires setting up a backup regime where an image backup is made of the OS partition on a periodic basis (every night is best). After the first backup (which is usually to a network UNC path), an operator can quickly recover a failed machine by simply inserting the disaster recovery CD and booting the failed machine.
You can restore to the original hard drive or a new disk right out of the box. No formatting other than a standard low-level disk format is required. Whatever was on the original disk, including multiple partitions will be restored exactly as it was last backed up (even Compaq hidden EISA partitions). This makes for fast diagnosis and recovery of bad hard drives.
The restore process approaches the maximum speed that the hardware configuration is capable of sustaining (tape drive, controller, disk drive, network path, etc.). After the restore process is complete, a reboot is all that is required to bring the operating system disk back to the state it was when backed up.
This process also allows an administrator to go to another machine with the same or similar configuration and perform a recovery, quickly turning the alternative machine into the identity of the failed machine.
Image backup is relevant even with dynamic mirroring and clustering! Dynamic mirroring copies corruption and file deletions and therefore automatically corrupts mirrored sets. A failed cluster node should also be recovered and brought back online ASAP or a single remaining active Cluster Server node can become a single point of failure.
Image backup is relevant even with file-by-file backups. While single files can be restored from image backups, the real power of image is quickly restoring a failed OS partition or replacing a bad hard drive. Since image backups can be made to UNC path and FTP/TSM devices, all machines can be protected regardless of whether tape is available or not. But image technology is not without limitations! An image backup is all or nothing – everything in a partition is backed up completely with no file selection logic or incremental or differential backup capability. Image backup is also a poor substitute for using file-by-file backups with API based database agents for backing up open and active relational databases like Microsoft Exchange and SQL.
Image backup is relevant even with RAID. Hardware-based RAID is transparent to image backup. While RAID insulates a system against a single disk drive failure, it cannot guard against registry corruption, deleted files, or a single point-of-hardware failure.
The combined attributes of file-by-file and image backup technology provide users with the safety and security of being able to completely control their emergency restore process and use the technology that best suits the situation. With UltraBac, users can combine both file and image backups or use them independently. This allows administrators to perform network backups of critical OS partitions using image technology and use regular file-by-file backup of all other partitions containing open and active relational database files like Exchange and SQL. This backup strategy provides for lightning-fast OS restores, guarantees the integrity of databases, and allows full, incremental, or differential backups to any media of choice.
Bottom line, image backup coupled with its universal boot CD represents the fastest and simplest method to recover from a failed operating system disk. Aside from speed and the resulting cost savings, image backup and restore technology completely eliminates the frustration of recovering a failed system.
Until servers and workstations have single point-of-failure potential totally eliminated, using image technology, with or without regular file-by-file backup, is a low cost option that can supplement dynamic mirroring, clustering, and RAID to further enhance system reliability and uptime.
Morgan Edwards, President & CEO, UltraBac Software