One of the most exciting — and at the same time, most frustrating — things about information technology is the rapidity with which changes occur and new hardware and software products emerge, with fantastic new "must have" features that make it easier for you and your employees to do your jobs.
There are two extreme approaches to upgrading:
- Some companies, especially those in the tech industry themselves, desire to be on the cutting edge and always have the latest and greatest. These early adopters toss out their old systems and replace them with the new as soon as it's available.
- Some companies, especially those in more "traditional" or conservative fields, go by the "if it's not broke, don't fix it" philosophy. They tend to hang onto their old tried and true technologies and upgrade only when forced to do so because a new feature set is absolutely required to get the job done or, more often, because the old machines finally break down and the currently available replacements don't support the old software.
These two strategies have something in common: they aren't really strategies at all. They're both reactive — the first driven by reaction to the marketing of new products, and the second driven by reaction to necessity. The smarter course is to develop a plan and timeline for upgrading, taking many factors into consideration — including scalability. A long-term plan must take into account future growth and changes anticipated by your company and the way those changes will affect your upgrade needs.
Categorize and prioritize
The first step in putting together your upgrade plan is to categorize and prioritize potential upgrades. Some categories of upgrade items include:
- Client computer systems
- Peripherals (monitors, scanners, printers)
- Client software
- Server hardware
- Server software
- Security devices (firewalls, biometric and smart card readers, IDS appliances)
- Network devices (routers, switches, network printers)
- Network infrastructure (cabling, Internet connection)
Of course, you don't have to upgrade all of the items in a particular category at the same time. For example, you might plan to upgrade the client computers in a particular department, such as the graphics department, based on those users' greater need for the features of new operating systems and applications. Or you might upgrade only certain servers, such as your mail servers, to take advantage of a new release of the mail server software that will make it easier to support your mobile users.
Thus you may want to create subcategories, such as "Client computers — accounting department," "Client computers — graphics department," "Client computers — upper management" and so forth. Similarly, software is a very broad category and you may upgrade operating systems and/or specific applications while continuing to use older versions of other applications.
Once you've identified the categories and subcategories, you can prioritize them based on the cost/benefits ratio of upgrading a particular category or subcategory. What items are serving as the bottlenecks that keep work from getting done as efficiently as it could with newer equipment and/or software? For example, if you're losing business from your e-commerce site because of the site's slow performance, speeding up the site is an upgrade priority. But you still must analyze the cause of the slow performance to determine whether you need to upgrade your Web server hardware, Web server software, or your network infrastructure (perhaps moving from a T-1 to a T-3 Internet connection, or spreading the load across a Web server farm instead of overloading a single server).
Of course, in real life your upgrade priorities may not always be based strictly on need. If the big boss wants the top of the line multi-processor workstation with 4 GB of memory and a high performance video card just to read email and compose the occasional Word document, there's a good chance he/she will get it. In general, though, don't over-upgrade. Plan to give your users the hardware and software that's required to most effectively perform their job tasks – and no more.
Upgrading some categories may be dependent on first upgrading other categories. For example, you may not be able to upgrade your productivity applications until you first upgrade the operating systems — and you might not be able to upgrade the operating systems until you've upgraded the hardware.
It can work the other way, too; if you upgrade the operating system, you might be forced to upgrade the productivity application because the old version doesn't run well (or at all) on the new operating system.
These dependencies affect your upgrade priorities and timelines.
Once you've decided to upgrade a particular category or subcategory, you shouldn't jump in feet first and roll out the upgrade to every system or device in that category. What if the upgrade causes major problems that make systems or the network unusable? The most prudent strategy is to roll out each upgrade in phases. Test it first in a non-production environment. This gives you a chance to work the bugs out without any impact on employee productivity.
Next, select a pilot group to test the upgrades in the production environment. If the new hardware or software entails a learning curve, roll it out first to power users, those who are more technically savvy and thus better able to handle the new way of doing things without overwhelming your support staff. Once they've mastered it, they'll serve as a resource for helping other users make the transition when you roll it out to the rest of the department or the rest of the company.
Keeping it scalable
Your upgrade plan should be set out in writing, and you should get input from different departments and different levels to help you create a plan that will create the least disruption and proceed smoothly. You'll need to know about any plans for expansion (geographic and in terms of personnel), so you can include the additional locations and/or users in the upgrade plan. Likewise, you'll need to know if there are restructuring, consolidation or personnel cuts in the company's immediate future. It would be a waste of time and money to upgrade systems that will sit idle a few months down the road.
Upgrading can be costly and traumatic, but sooner or later it's inevitable. Proper planning, with scalability in mind, can make the difference between a smooth deployment of nifty new technologies and an upgrade disaster.
Debra Littlejohn Shinder, MCSE, MVP is a technology consultant, trainer, and writer who has authored a number of books on computer operating systems, networking, and security. Deb is a tech editor, developmental editor, and contributor to over 20 additional books on subjects such as the Windows 2000 and Windows 2003 MCSE exams, CompTIA Security+ exam, and TruSecure's ICSA certification.