Enterprises understand the importance of business continuity because downtime costs lots of money. With downtime costing the average enterprise more than $1 million per hour, the return on investment (ROI) for back-up protection can be counted in minutes and seconds, not in hours or days.
Make no mistake about it: Downtime can be hazardous to your organization’s financial health, even if you’re not running a Fortune 500 company. In fact, small- to mid-size businesses are even more at risk.
Gartner Inc., a leading market research and consulting firm, estimates that two out of five companies that experience a disaster will go out of business within five years as a result of the event. This U.S. Bureau of Labor Statistics is even more pessimistic; the government claims that 93 percent of all companies that experience "significant data loss" are out of business within five years.
Why is this when most companies say that make some sort of backup for their servers? The reason is painfully clear: backing up is one thing, but being able to restore your systems is quite another. In fact, because "disasters" vary in intensity and manageability, often the data that is impacted by the disaster has not been backed up at all.
According to IDC, 60 percent of corporate data resides on laptops and desktop PCs. Moreover, an employee without a desktop has severely curtailed productivity while IT is restoring the system – and just imagine a situation where a virus infects an entire network, affecting multiple employees. Organizations cannot afford to lose data, much less have downtime. It costs too much.
According to Gartner, in a company with 200 employees who access three LANs with 98 percent accessibility, that 2 percent downtime translates into 31,200 hours of lost work hours per year. Put another way, that 31,200 hours translates into 15 man-years. That means that a company with 200 employees is paying the equivalent of 7.5 percent of its employee salaries to people sitting around waiting for the LAN to come back up. That’s an awful lot of non-productive time.
How can a small business protect itself from massive downtime and wasted expenses?
There are essentially two types of backups: file-based and image-based. A file-based backup is akin to using Microsoft Windows’ copy command to move files from your hard disk to another location, generally another disk or tape, but sometimes a DVD or CD. While this command is very effective for copying user files, it is ineffective when copying open Windows files. Generally, to use a file-based back-up product, it is necessary to boot to DOS first. From DOS, you can copy all of your files. However, when you’re in DOS on a server or a workstation that system is locked; the user is effectively shut down until the process is complete. In a 24x7 environment, that simply is not a cost-efficient solution.
The second type of backup is creating an image on the hard disk. Think of it as a snapshot in time. The most effective image-based products can create this snapshot without leaving Windows. As a result, your employees remain productive, and the image is created in the background.
Incidentally, images can be made of either servers or workstations and can be scheduled. But be aware – not all imaging solutions can create an image over a LAN to a different network domain. If yours can’t, it means you’ll have to purchase and maintain management consoles on every network domain in the company.
Let’s look at a few back-up scenarios you might face.
Scenario 1: Bare-Metal Restore
A business has lost an entire server. The server needs to be replaced with a new box. This usually involves building a system from scratch — an unformatted hard drive without an operating system is installed, and everything is installed from scratch. In the worst-case scenario, the IT administrator not only has to reinstall all the system’s software, but it has to be installed on dissimilar hardware.
Some of today’s imaging software allows the IT manager to restore from an image, but you may be forced to first strip out the security identifier (SID) number and all network information, including all user names, rights, and the like. Once the image is restored, it can take hours to reconfigure the server.
If the server’s identity is maintained, the IT manager needs only to add new hardware drivers when prompted. When the system reboots, it will be live on the network with all of its identity and configurations intact.
Scenario 2: Backup of Windows Data Center
A business is using 10 servers all running Windows. The servers are split into two data centers, with each location having five servers. The company can backup and restore all 10 servers from the system administrator’s desk, regardless of the domain, simply by scheduling or manually backing up the servers from a management console. This console can reside on any system – even a laptop. Only a small agent needs to be installed on each server.
Scenario 3: Backup of an Entire Network
A business is running five servers in a data center and has 200 employees, each with a desktop PC or laptop. The business is looking for a solution that can backup and restore all servers and desktops from one system, as one IT staff member is responsible for ensuring that business can continue, even in the event of system failure. The business would purchase a software license for each server and workstation. The IT staff member installs the management console on his desktop PC or laptop and remotely installs agents on all 205 systems in the company.
Remember, your backup is worthless if you can’t restore it.
Stephen Lawton is senior director of strategic marketing for Acronis, Inc., a leading developer of storage management software based in Burlington, Mass.
"Appeared in DRJ's Summer 2007 Issue"