• United States

Enough with the ‘lifecycle management,’ already

Jun 30, 20043 mins
Data Center

* Why ‘lifecycle management’ is a bunch of hooey

Today’s column will focus on one of my pet peeves: the misuse of illustrations in an attempt to explain and organize management products. A blatant example of this, and one that drives me up a wall is the use of “Lifecycle Management” to explain the offerings of server and desktop management vendors, particularly software distribution vendors.

I’ll scream the next time I see another lifecycle management slide in a server or desktop management vendor’s presentation. Nearly all of them are using this illustration to depict what they sell.

First of all, the term “lifecycle” usually implies a logical and predictable progression of stages with a defined beginning and an end. For example, a person is born and enters infancy, then becomes a toddler, and then on to adolescence and puberty, then fast forward to middle age, on to old age and then the end of life. All of these stages are very predictable, and they are logical progressions. You simply can’t jump back to adolescence from being a middle-ager.

There is a legitimate case for using “lifecycle management” when describing asset management from a financial perspective. The asset is procured, tracked through its useful life in the company, and then disposed of when obsolete.

However, there’s nothing logical nor predictable about server or desktop management. The servers and desktops are procured and then deployed, and these systems are also retired from service. But it is at that point that the predictability stops. During their useful life, these systems can be repurposed to another user or for another use, which means redeploying the appropriate software for its new use. Software patching occurs on an ongoing basis, as well as periodic software upgrades. Throw into the mix the possibility of requiring software migration. Users may also add their own applications, or IT may add new applications. For servers, IT may decide to do server consolidations, which may drastically change the configuration and use of the server.

Then there is another emerging approach that is not yet widely used but is being talked about by vendors: automated provisioning. If automated provisioning takes hold, then all bets are off for any kind of semblance to a lifecycle. Servers could be swapped into and out of server pools, temporarily repurposed for an application, then swapped out and re-provisioned for another application.

None of these functions are performed at predictable stages; they’re just done on as needed. The servers and desktops don’t have “lives”; they are boxes that are used for whatever the current need is.

Why are the server and desktop vendors all using “lifecycle management” to illustrate an environment that doesn’t map to a lifecycle at all? As I see it, there are two possibilities. Either the vendors don’t understand the dynamic nature of the server and desktop environments that ITers have to deal with, or it’s a marketing ploy to convey broad and full management of the server and desktop environment, as well as trying to simplify the explanation of how their tools fit together. It is more simply illustrated but clearly not representative of the server and desktop management environment. In my view, neither of these reasons justifies the use of “lifecycle management” in this context.

So the next time you see a server or desktop management vendor touting “lifecycle management,” please do me a favor and ask them to stop using it. If we stick together, we can eradicate this menace from our world – well, at least we’ll have one less irritant to put up with.