Databus Issue: 2009 1 03/31/2008
Best Practices
Phil Scrivano Vice President of Customer Services
PDF
IT reality dictates two things: all technology runs on electricity and someday all technology will stop working. Today’s competing reality is, in the event of failure, users feel they cannot work without their technology. In the early days of putting computers in classrooms, if there was a failure we often took a week to set up a replacement or repair a power supply on an Apple IIe. Teachers did not seem to grumble much about losing the ability to have students use LegoWriter. Now the computer and the connection to the WAN and the Internet are almost an essential part of the classroom. I use the words “almost an essential part” to remind teachers that a good teacher can still teach without technology. These days, school district IT departments are amazingly fast at getting things working—but the more efficient we are with delivering technology, the less tolerant of downtime our users have become. Consider as an example that your users can always state the date, time, and duration of the last time the Internet connection went down.
Minimizing downtime is a mandate, so consider standardizing your technology. The more you are able to standardize technology, the faster you can get it back running when you have failure. As a technology director in the 90s, I learned that at the beginning of each new budget year my boss could be convinced of the need for a spare-parts supply order. I would keep a list year to year of what failed and what it took to get it running again. The most common items were a selection of hard drives for most computers, a selection of computer power supplies, internal cooling fans, optical drives, UPS batteries, fiber patch cables, various lengths Cat 5 cables, common projector bulbs, and at least one common 24 port switch. By planning ahead in this fashion, there are many costs and public relations benefits.
A related topic when planning for failure is to budget year to year for the largest single points of failure on your network. If it is a core switch, firewall, or server hardware, try to budget over a five-year period for replacement. Business officials are much more willing to work with you in this manner versus an unplanned major expense when the device fails.
