Online Backup Services: Page 2 of 22

Even if you can swing these expenses, you'll still have to find personnel at remote locations who are capable of ensuring that backups are properly handled. That's a huge challenge: The average human is ill-equipped to review a typical error log, much less tackle a jammed tape drive or server rack of flashing error messages.

This is where online data backup services can save your bacon. After initial configuration, the service monitors your servers' backup status while invisibly handling garden-variety backroom problems. Failed backups and tape-drive implosions can become a thing of the past, and you'll no longer lose sleep over hardware contingencies for data growth or spikes--that's now your service provider's problem. Even more alluring, you can usually negotiate lower rates for higher-volume storage as your data needs grow.

We found substantial variations in the cost and features of online server backups, the main sticking points being storage volume and length of retention (for a side-by-side comparison of the features offered, go to www.nwc. com/go/SE1004RD1.jhtml). Most online services log each incremental backup as a restoration point, and the retention scheme (and cost) is based on the number of iterations available for restoration. For example, on a service with 10 retention levels, files marked for backup remain on the service indefinitely, but only 10 historical versions are available for restoration. A file deleted from that data set will remain available for restore until the 11th incremental backup is made.

As a rule, cost is based on the amount of data being stored and the services' retention scheme. The first full backup is usually the most bandwidth-consuming part of the process; for ongoing incremental backups, most services use byte- or block-level file analysis as well as data compression to reduce the size of subsequent transmissions dramatically.