|
 |
|
General Information
|
Add To My Personal Library |
December 8, 2006
• Vol.28 Issue 49
Page(s) 11 in print issue
|
Manage Your Assets, Manage Your Budget
Equipment Life Cycle Tips From The Pros
|
 Just like humans, servers, notebooks, and printers grow old and die. But you can maximize their useful life to save money—and believe it or not, simply using them longer may not be the best way.
“Holding on to old equipment can appear to reduce IT spending,” says Darin Stahl of the Info-Tech Research Group, “but doing so may actually cost you more money in the long run.” (Disclaimer: Info-Tech analysts write a regular column for Processor.) He notes that increased support costs, upgrades, replacement parts, and downtime will drain dollars from your budget. And there are other, subtler costs to keeping equipment longer than you should.
First among these is the cost of making new software play with old hardware. “Consider the delay that this could create for project implementation and the costs associated with stretching project timelines and paying idle project members,” says Stahl.
So how do you maximize your equipment’s life cycle, the better to minimize your budget? We’ve gathered some expert tips below.
 Start With A Plan
Gartner analyst Patricia Adams believes that small to midsized companies should manage their equipment’s life cycle with a careful, well-conceived plan, one that covers each stage in the process: requisition, approval, and acquisition; receipt and verification; deployment and management; reuse or retirement; and end-of-life and disposal.
“By developing a master plan for how and what will happen to an asset throughout its life cycle,” says Adams, “these processes will improve IT costs, customer satisfaction, and also improve business performance as a result of that.”
A good plan can help you avoid costly errors such as unused warranties. “If you order a PC and it sits in a warehouse or a stockroom for three months, instead of using that for four years, you’re actually ending up using that for three years and nine months because the warranty has been in effect,” says Adams. As a best practice, she recommends that companies track assets from the moment they’re requisitioned.
Avoid Support Problems
According to Info-Tech’s Stahl, old hardware is a risk if the vendor or manufacturer no longer supports it. He suggests you act now to avoid problems. “As a start, get all the documentation possible for older systems and the associated procedures,” he says. “Don’t rely on the legacy knowledge of a few workers as a support network.”
Stahl also recommends you keep a spare parts supply, keep up preventive maintenance, and keep a close eye on older gear. “Analyze support tickets in a manner that allows you to identify patterns of failure and single out faulty hardware,” he says. An ounce of prevention is more than worth an all-nighter when a server goes down.
Be Precise
Do you know how many assets your data center has? If not—and if you don’t know how to find the answer, save for counting up servers and notebooks and routers by hand—then let software lend you some precision, the better to maximize your equipment’s useful life.
EAM (enterprise asset management) software will not only help you craft a strong asset management plan but also automate it. Some programs even automate asset detection itself, scanning your network to determine every endpoint, server, router, printer, fax, and handheld you have, not to mention high-end software such as CRM and ERP apps, and even licenses.
The result? Programs such as Numara's Track-It! (www.numarasoftware.com), PerfCap’s FindIT (www.perfcap.com), and Liberty Street Software's AssetManage (www.assetmanagesoftware.com) can make it easy to track a thousand devices, knowing the precise point to send each one to the scrap heap or cycle it to a different use. (For more on cycling, see the “The Easiest Way To Save Money” sidebar.)
Get The Most Out Of Servers
Speaking of servers, get the most out of the ones you’ve got, before you buy more. Info-Tech’s Stahl says the best way to make end-of-life server decisions is to monitor a box over time, looking at “storage throughput, network bandwidth, and CPU utilization.” Problems such as excessive disk activity might point to a need for more RAM—a cheap upgrade compared to a full replacement. And tools such as HBAs (Host Bus Adapters) and TOEs (TCP Offload Engines) can reduce the load on CPUs.
Some servers, of course, have simply met their ends. “If the server CPU utilization is consistently over 80%, or reaches 100% even for short periods, then this machine has reached its capacity for this particular application,” says Stahl. 
by David Garrett
The Easiest Way To Save Money
Here’s an easy way to pinch pennies through life cycle mastery: Use what Info-Tech’s Darin Stahl calls a “cascading refresh.”
“New servers can push midlife servers down, replacing old servers in a cascade effect,” he says. The new servers run mission-critical services. Midlife servers can be used for important but nonvital functions, such as static Web sites or once-a-month billing, and old servers can be ditched.
“Replacing the oldest servers in the data center with midlife units allows the server investment to be amortized over longer periods of time,” says Stahl. It not only saves money on your balance sheet but lets you intelligently match your equipment’s power to its business use, changing the equation over time. |
|
|
