Computer Security Article

Protecting the Enterprise:
Systems in Reserve

Author: Michael Miora
Published in
Carolina Computer News, May, 1997

Imagine a Small Failure

Imagine this. You are at home and about to do some work on your computer. You turn it on and let your mind wander while waiting for your operating system to start. Moments later, you see this message on the screen:

HARD DISK ERROR - PRESS F1 TO CONTINUE

Your heart stops. You press F1, only to see your system halt waiting for your floppy disk drive to supply a system. You press reset. You try booting off of a floppy and reading the hard drive. You do this many times. You stare blankly at your screen, desperately hoping the hard disk will repair itself. Maybe it is a virus playing tricks on you? Maybe the drive is stuck and a short, gentle tap on the side will start it. Okay, maybe a little harder. To no avail - the disk is dead.

Disks are getting larger in capacity and lower in cost with every passing week. Unless you bought your computer in the last few months, you can probably replace that worn out old drive with a new one that is double the size and twice as fast for half what you expect to pay. No problem, right? Just go buy a disk and restore your backups. You'll be back up and running in a few hours. No backups, you say? Now that could be a problem.

Lets suppose you are a heavy duty computer user. Your financial information is on the computer. You track your bank balances and pay your bills using Quicken. You were going to pay the bills today, since they are already late, and you were going to pay as many as you could before running out of checking account balance. Now you don't know how much you have. And you don't have any written records.

You also suddenly realize that all of your phone numbers and addresses are gone. You used Schedule+ to keep things organized. Now the numbers are gone and you'll need to recompile all those phone numbers and addresses. You never did print that list.

Your friend from across the country said he would send you email with his flight information. He's arriving tonight. You can't remember his phone number and, of course, you can't check your email.

This situation can be prevented with a small amount of advance planning and a few hundred dollars to buy a new disk. This article is not, however, about scaring people into backing up their computers. If I succeed at that, then so much the better, but that is not the primary point. This article is about companies with networks of computers.

Imagine a Large Failure

Imagine this. Your office computers are damaged or missing. Such things happen all the time: fire, water damage from rain or broken sprinklers, and burglaries, just to name a few causes. Your office now cannot function. Just as above, financial information was on the network. Communications was on the network. Memos, documents, and miscellaneous other work products were also on the network. This loss can be serious. Of course, many companies do make regular backups. Many do not, but that is fortunately a shrinking group.

In the scenario at the beginning of the article, a simple replacement disk and backup restore would solve the problem. In this scenario, however, the recovery is significantly more difficult. Many offices use multiple servers. Restoring the backups requires several servers and multiple workstations, along with a place to put them. If the business needs to function immediately, then the 1 to 2 day delay associated with replacing systems, restoring gigabytes of data, and configuring workstations may be unacceptable.

Reserve Systems

Enter the reserve system. This is not a backup copy of the regular network. It is a small computer along with selected data to perform key functions for a short time. A small PC can today perform large and complex tasks if the software and data is available. There are many instances where individual microcomputers substituted for large computers for short periods of time.

With good advance planning and some creativity, many options are available. For example, a facility for a California company warehoused all products produced in the Western United States for world wide distribution. This warehouse was semi-automated: fork lifts and other moving equipment were operated by workers, but locations were encoded using bar codes and instructions were received in real time from an on-line minicomputer system. The basic computer (located at another, nearby facility) system provided end-to-end processing, including order entry, shipping/receiving, invoicing, warehouse management, and collections.

If this system were damaged, all shipping would cease. The company decided that public relations, financial, and legal ramifications precluded this function from being interrupted for more than 24 hours. The minicomputer, however, could not be reliably restored and brought on-line at a hot site in less than 36 hours without incurring very high expenses.

The company built a Reserve System. Using commercially available data base software engines along with friendly front ends, the warehouse data base was parsed into two parts: the functions and data that must be on-line within 24 hours, and the functions and data that could wait 3-4 days.

A "Spartan" version of the software was built to run on a PC. This version of the warehouse software performed only those two critical functions deemed necessary: receive, put away, find and ship product. Selected data from the minicomputer system was downloaded to a remote PC nightly. That PC was located at the warehouse, and connected to a wireless LAN installed specifically for this function. Fork lifts and other equipment used their existing on-board or hand-held equipment to access the PC using the wireless LAN and some special purpose signal conditioning equipment.

This Reserve System performed the critical functions that enabled warehouse personnel to operate at 50% capacity compared to normal operations -- but the functions were performed and the company delivered its products.

Continuation versus Recovery

The goal of reserve systems is to continue operations as well as possible. If you had backed up your Quicken data on to floppy, then when your hard disk crashed, you could have gone to a friends house with your Quicken installation CD and your backup disk and you would have been able to continue. This is business continuation.

Recovery is the rebuilding of full operating capability. That takes longer, and speeding the process significantly raises the costs. Reserve systems enable an organization to continue functioning while recovery systems in an orderly manner.

Reserve systems should be identified, built and maintained as part of every company's Disaster Recovery Plan. Now, go back up your hard disk.

Articles


Updated Spring, 2002 by webloke © Stephen Cobb
Some article content reprinted by permission.
Article content copyright named author(s).