Try these beans: Business Service Management In recent weeks we’ve received plenty of meaty responses in roughly equal measures (positive and negative) to the questions posed in this workshop. For many, service management is seen as simply doing the job properly, regardless of the tools / politics / obtrusive management and so on. Others do what they do but really don’t …
OK I admit it, I work for a company that does agentless IT inventories but honestly, things will be much better the day we stop lying to ourselves and quit pretending that we are in full control of what's happening in our IT shops. You know that sh*t is happening, at least try to know what kind of sh*t is coming your way!
Ah, yes. The elusive "End-user experience". Synthetic transactions tests? embedded metric collection within the app like A.R.M.? External passive monitoring? Good luck trying to explain the nuances, pros and cons of each to your executive management, and also that they'll probably need to buy something, but it won't necessarily provide Business Services Management.
Then they'll look at you and say, "but we're deploying everything in BPEL now!", and you'll sigh.
If you're a typical shop you have multiple tools from multiple vendors collecting most of the underlying data necessary to accomplish BSM, but where to do the management? Most BSM solutions provided by vendors are part of a larger suite (usually purchased for the best-of-breed, point solution "X" but hey, you get this other cool stuff at a discount!) and integrate with your other vendor software poorly or not at all, leaving important data out of the overall service representation. Several products claim to provide BSM capabilities via some sort of dashboard, but upon closer inspection don't really allow weighting of metrics and application of differing algorithms against underlying metrics to derive service state, critical in large-scale clustered environments.
Implementation is a labor intensive process as you begin to normalize the underlying data and represent these metrics as part of the application architecture, in some cases using other tools and system forensics to complete the picture for legacy apps where the institutional knowledge has long since left the building.
Finally, unless your organization has a fairly solid ITIL implementation, you can expect to spend the remainder of your time maintaining these business views for applications which change often. A good ITIL implementation will allow you to automatically populate and update the business models based on attributes and information defined within you change management/change process system.
Oh, did I mention patience? You'll need lots of that. :)
SWS
IT is maintenance. Overhead. A cost center.
Think janitorial, or grounds/building maintenance, only slightly more spendy.
IT is needed in this current era, and IT costs money (sometimes LOTS of money), but in reality it doesn't, in fact, actually MAKE money in and of it's own, any more than cube farms full of people using 13-column pads and 10-key calculators did in the 1960s and '70s.
It's up to Management to figure out the direction any given company is going ... however, Management rarely looks into the details of how the plumbing works, or who is cleaning the windows or trimming the shrubbery. That is left up to the maintenance people.
Unfortunately, computers (and the use of computers), gets very expensive when you head into the several hundreds of seats range ... and extremely expensive when your eclipse 1K seats ... I won't get into the costs & complexities of 100K+ desktops spread world-wide. That's a lot of money being spent on what traditional management sees as "just maintenance" ... So traditional management thinks that it must be something more than maintenance. IT isn't.
Any company with half an ounce of common sense in this era will have both a "technical" track and a "managerial" track ... In this scenario, management does management stuff, and a more technically inclined person with managerial ability, hopefully sitting at the Board level, manages the technical side of things that most management quite frankly aren't properly equipped to understand. This person's duties should include hiring, firing, and promoting of more technologically capable employees (or, in larger companies, appointing staff to do same).
Traditional management's feudal derived mindset doesn't work with IT. IT changes too fast. Traditional management can't cope with fast changes. Trying to define IT in traditional MBA terms is, in my mind, an exercise in futility.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
