Approach IT lifecycle management with this question in mind: By extending the life of underperforming servers,...
are you hurting the data center in the long run?
Take a look around your data center. All that equipment sitting there. How much did you pay for all your IT assets? How much is that stuff now worth?
Now the big question: How well is that equipment managing workloads for the business?
IT equipment ages rapidly -- not in the sense that it fails or misbehaves, but in that new, faster, more energy-efficient equipment comes to market on a seemingly constant basis. Some of the equipment in your data center that's three or more years old may not be up to the job any longer. If you have been squeezing IT assets by extending their lifecycles over a five-year period, that equipment could actually cost you more in business terms than the price of new equipment.
Many organizations have little real understanding of what is in their data center, never mind carry out real IT lifecycle management (ITLM) processes around the equipment. ITLM can not only save money for an organization, but it can also ensure that the most optimal IT platform is in place on an ongoing basis.
ITLM requires that you find out and log exactly what equipment assets are in the data center. Use an IT asset management tool. Many organizations carry out rudimentary asset management with spreadsheets with serial numbers and dates of purchase and delivery to keep track of items, but these systems rapidly become out-of-date as equipment is updated with new components. Things can fall through the cracks -- e.g., when hardware fails and is removed from the premises but not from the spreadsheet, or when systems engineers bring in new equipment and fail to enter it into the spreadsheet.
What's needed are automated tools that can carry out a sweep of the entire network and identify not only the endpoints, but also deeper details about vendors and a granular breakdown of components, such as attached-storage devices, graphics cards and network interface cards (NICs). Add in information at a software level, such as operating system, applications housed in servers, software version levels and what patches were applied. That's a pretty full list of assets.
How to value IT assets
Once you really know what systems are in place, go deeper. For example, maintenance schedules can be added with triggers to fire off tickets into ITIL service management systems, providing full work dockets and details of the equipment's exact location, so that engineers work on the right equipment with minimum time wasted. More information means that admins can ensure workloads move off of systems that need to be taken down for maintenance, so as to prevent downtime to users.
But the real value comes in being able to apply value economics -- a way of ensuring that an IT platform is at the optimum level of performance at all times based on the business and economic value of the underlying assets -- to the equipment.
IT hardware items have three basic values. Once the item, for example a server, has been purchased, it will be placed "on book" and will be depreciated at whatever rate the organization deems to be suitable -- generally three to five years with a straight-line depreciation.
This item has an intrinsic value -- the amount that someone would be willing to pay for the equipment on the secondhand market. This is not straight line -- there is a rapid depreciation in intrinsic value from initial purchase, flattening out somewhat over time.
Then there is the data value of the equipment. Most servers will have direct-attached storage; storage systems are full of data; routers, switches, other network devices, and even printers tend to have small storage capabilities that can hold data with corporate value, such as username/password pairs.
The final component to take into account is business value. This is the one that is most dependent on the aging equipment. To start, a brand-new piece of equipment has little direct business value -- it only starts to provide this value once implemented and supporting workloads. As the data value increases, the business value will also increase -- but the relative decrease in the speed and capabilities of the equipment against newer equipment begins to dilute the overall value.
The trick is to be able to identify exactly where the peak of business value is and replace the equipment at the point where this begins to fall off.
At the point of replacement, the intrinsic value can be realized and offset against the purchase of the new equipment, providing some level of financing towards constant equipment refresh.
There are lots of areas that need to be borne in mind, though, around selling old equipment -- particularly when it comes to data security. The data value of the equipment will be high to others as well as to the business. The data will have to be destroyed before selling the equipment. Depending on the organization's own risk profile, this may be through the use of secure information deletion or may be through the physical destruction of hard disk drives -- lowering the value of the remaining equipment, but maintaining corporate security. However, even the detritus of a destroyed disk drive has value -- it contains copper, gold and other metals worth recovering by commercial organizations specializing in this field.
Data center lifecycle management should be in place for any organization as it tries to optimize asset investments, even if doesn't go as far as selling off old equipment to help fund new replacements. However, full lifecycle management has financial and business benefits that are very powerful -- organizations should look into how such an approach can be implemented.
Clive Longbottom is the co-founder and service director of IT research and analysis firm Quocirca, based in the U.K. Longbottom has more than 15 years of experience in the field. With a background in chemical engineering, he's worked on automation, control of hazardous substances, document management and knowledge management projects.