This article is also available as a PDF download.
2006 has sped by, leaving us with little time to ponder the approach of 2007. Massive upheavals in the OS, software, and processor spaces have left us with new, and radically more expensive, distributed architectures to consider. Meanwhile, our operational and support environments continue to decay under the weight of corporate neglect, since we have trouble showing ourselves as anything other than an expense.
The following observations are intended to help keep you grounded in the basics while dealing with the challenges of the new year.
#1: Bigger is not better
We can now buy a blade enclosure with 86 cores in it. A full rack could hold more processing power than some server rooms we were thrilled with in the late 90s. We can hurl memory at problems for pennies on the MB, and media access keeps scaling up to ludicrous speed.
That's all great, but it doesn't relieve us from our duties. The ability to mask problems by overbuilding hardware does not mean we've solved the issue—it just hides the problem for a time. Similarly, the ability to build a data center in a rack doesn't mean that doing so is a good idea.
#2: New does not mean stable
It usually means the opposite. We call the adoption of radical new ideas "the bleeding edge" for a reason. However, in the coming year, many of us will face the terrible challenge of choosing between remaining where we are, with our currently stable support structures, and moving to new platforms where our support teams will have to retool their knowledge base.
New environments do not make for stable operations and support. Before jumping headlong into a new operating environment, determine whether the business can endure the turmoil.
#3: Virtual devices still reside on physical hardware
Virtualization has finally come into its own. We can make one server look like five, share limited resources, and create a "logical" environment that only vaguely resembles the physical one.
Just because we can doesn't mean we should. Always remember that at the end of the day, all those virtual servers live on a physical box somewhere. That box needs the same maintenance and support it always did and more, now that we've added yet more layers of complexity.
#4: Follow-though matters as much as execution
As project portfolios grow and management takes control of all three sides of project management's iron triangle, there never seems to be enough time to follow though with activities. We get one thing done, then jump to the next three, hoping that somewhere in all the chaos a real change took hold.
Rather than sprinting from failure to failure, take a few moments to follow up on the activities of the past. You may well discover that the thing you thought was a failure can, with just a little nudge in the right direction, turn into a success.
#5: One size fits all never fits anyone at all
We all know this, but it bears repeating. Boilerplates, best practices, and the fabled "industry leaders" all suggest that one solution will fit every situation. We just have to ram it into our environment and everything will magically transform into rainbows and bluebirds.
Every IT environment represents a unique collection of people, processes, and technology. Best practices generate terrible distortions in the environment when we try to implement them without first accounting for the people and how they actually get things done.
#6: It's better to invest time than to spend it
Stop trying to manage time. Time is not something we spend; it's not a resource we can horde or a precious jewel to display. Time is life; our life, our energy, our passion as it expresses itself in the world.
Instead of treating time as a thing to spend, we should consider our actions in terms of the kind of life they build for ourselves and others. Every day we face choices about how we shape the world that our co-workers, employees, and managers live in. Once we accept that, we can create genuine change.
#7: Bolster morale though consistent measurement
It is easy, even tempting, to tell people that we will know success when we see it. I hear it all the time in my consulting practice with predictable results. People who do not know how they will be measured perform poorly. Those who know that the standards change from one person to another (and depending on the manager's stress level) simply assume success is not possible.
We have to tell people what we want and expect from them. Otherwise, how can they possibly perform?
#8: Failure is always an option, so fail early and often
Motivational speakers aside, most of us know that failure happens. Success is difficult to achieve and fleeting in the world of IT. It's not always fun, and frankly it's pretty unpleasant, to go from meeting to meeting in which executives chant "failure is not an option."
We should embrace failure. In fact, we can turn this reality of our work lives to our advantage. If we cause doomed projects to fail early in their lifecycle, we can free up resources to deal with projects that have a chance of success. It isn't a terribly popular approach, but it does create the opportunity for success. It also exposes all the risks we take, daily, in the name of making a profit.
#9: It's important to appreciate the difference between a target and a deadline
A target is just that—something we aim for. If we miss the target, odds are good someone will be embarrassed but nothing overly bad will happen. Targets generally come from corporate political discussions and seem terribly important at the time.
A deadline, though, carries some kind of financial or legal consequence for the company. Failure to meet a deadline can land the team, the department, or the corporation in some kind of serious jeopardy.
Targets we can miss from time to time. Deadlines we should take very seriously.
And there we have it. Nine things we can focus on to make 2007 a better year than 2006, even in the face of the turmoil ahead. Good luck and have a Happy New Year!