A laptop with two PC card NICs and the right operating system can route between two networks. It will be slow, but it will work. And while it is routing packets you have bought a little time to get the next day warranty router in to replace the original.
Can’t afford redundant fiber and managed switches in case of a fiber cut or other campus network failure? How about a phone line in the remote building and a modem to dial in and connect to the required systems? Wireless phones are fantastic when the land lines fail, but they are also well suited to relaying messages to someone who cannot receive e-mail at a remote location until service is restored. The person at the front end can read off the e-mail and then respond on behalf of the recipient.
The point is not to do everything on the cheap, but rather to take responsibility for making sure your company can continue to operate even when you do not have the budget to provide everything you would like. It’s irresponsible to throw up one’s hands and blame the budget. Solutions must be found, money or no money.
Another critical concern is who needs what, and when? Work with the various departments to determine exactly what processes must take place for the company to do business, and exactly who must have access to what information.
A mistake made by many IT managers is to decide for themselves what systems, data, and processes are critical to the organization. This decision should be left to the managers and supervisors on the front lines, not to IT. Enlist their help by asking questions like, “If this system were down for two days, how would you continue to operate? What could we do, either now or then, to make that process easier?”
There are two key advantages to this approach. First, by working directly with the people who are served by the IT resources, the IT manager is certain to focus on the systems most vital to the operations of the company. Second, the managers and supervisors at the front lines may have other ideas or alternatives available in the event of a system failure. If this is the case, the IT manager will not need to focus as much time and resources to provide a solution for that system. This way, the time and resources available to the IT manager can be directed only to the most vulnerable areas.
Besides the managers and supervisors around the organization, your best ally in reducing systems continuity costs may be your vendors. If systems continuity is not an option due to budget constraints, what about reduced recovery windows? Would a vendor be willing to loan or rent you a system on very short notice, even after hours, in an emergency? If so, acquire the appropriate guarantees, requirements, and contact numbers to make it happen today.
Consider this option when budgeting projects for the coming year. Rather than negotiate as far down on price as you can, work on negotiating such emergency services into your deal with the vendor. Communications companies may be willing to guarantee better SLAs in exchange for a slightly higher price than you would have paid otherwise, and product vendors may be willing to provide after hours emergency support in exchange for more of your business. If your vendors are not willing, or do not have the resources to do so, you may want to shop around.
Every business will have different needs, and every need will have different alternatives. The examples outlined here are only examples, and are designed solely to start the thought process. If there are systems that cannot be down, list each one on a sheet of paper and work diligently until you have found a solution for each.
Of course there are times when real-time active redundancy, and the expense that goes with it, is the only choice. A company that stands to lose $1,000 per minute when its Web server is offline will have to have the best in active redundancy. But if the Web server is generating $1,000 per minute, that company can probably afford it. For the rest of us, there are alternatives. We just have to knuckle down and make it happen.
J. David Harper has been managing networks for 14 years and was a speaker on disaster recovery at the 2003 CAMUS International conference. He is currently the network communications manager for a global manufacturing company based in Central Texas.