Data center migration day is a great excuse for getting rid of those things that just seem to be hanging on and taking up space. No, not your co-workers, but the old hardware, apps and data that no longer serve a purpose.
In this fifth article in a six-part series, Logicalis technologist Brett Anderson helps you figure out a system for sorting the chaff from the wheat when it comes to losing the tech trash in your data center.
“What if we have to go back to it?”
This haunting question, typically asked at the beginning of each new wave of change that sweeps through IT, has been used to justify keeping all kinds of data center detritus: old servers and workstations, printers, modems, fax machines; the last few versions of the database upgraded six months ago, and backup copies of the test, dev, pre-production, production and maybe even post-production instances of the application that “guy who worked here before” developed.
Having spent entirely too many late nights and stress-filled, long weekends scrambling to recover from one crisis or another, most IT professionals have a hard time letting go of technology that once worked. A cautious nature is an admirable characteristic in a technologist, but it needs to be balanced by a willingness to let go of the past and the confidence to commit to the future.
As more IT departments plan their move to a converged infrastructure, they need to take a very critical look around them and decide what they actually need to take and what they are never going back to.
Whether you are moving to a new data center or upgrading-in-place, it helps to prepare for a migration to converged infrastructure the way you would prepare for a move to a new home. I recommend parting with the stuff that causes the least separation anxiety: the old time-tracking system that is still running but hasn’t been used in years. Since you went to Internet-based faxes a year ago. Do you still need the old fax modem server? What about all the old POTS lines that no one cancelled?
Making decisions about leftovers like these is relatively easy and builds confidence for dealing with the other stuff you find around that requires more thought.
The ‘Dark Side’
Virtualization has made it possible to consolidate the physical server sprawl that threatened to overtake many data centers. The dark side of virtualization, however, is that you can spin out applications and virtual servers without even having to buy more hardware. The result in many data centers is rapidly escalating virtual server sprawl. This sprawl may not occupy as much floor space as physical server sprawl, but running all those VMs does consume processing power, memory, and storage, and generally clutters up the very infrastructure that you are trying to simplify and optimize.
Before you migrate anything to your new environment, we recommend conducting a thorough application assessment to determine which applications you need and which you don’t. This is not something the IT department can do on its own. For the assessment to be effective you need to involve the business side of your organization and find out what is and is not actually working for them.
The good news here is that if you are serious about implementing a converged infrastructure and providing IT as a service, you are going to need to communicate much more openly with the business leaders in your organization anyway. Use the application assessment as an opportunity to enhance the lines of communications and to build trust.
Sometimes what looks like junk turns out to be frighteningly valuable. During an audit for a client awhile back we uncovered an old Sun server that was still running. Only two of its disk drives were functioning, and no one was sure what they contained. It turned out that the faltering drives provided critical data to an extremely important but small and static application and, had one more disk failed, the IT department would have become painfully aware of the amount of risk it had been unknowingly taking. We were able to remediate the immediate need, but more importantly, we got the application on the planning board for review and replacement — in a controlled, manageable fashion.
Just because something is old doesn’t mean you don’t need it. But, if it does have value, it needs to be on a platform that you can monitor and manage into the future. It also goes without saying that, if you don’t know where (or what) something is, you can’t protect it.
The aging process is particularly unrelenting in tape libraries, for example. It becomes so automatic to spin data off to tape and file it in the tape library that few stop to realize that tapes have a limited shelf life. During backup audits we have found whole sections of tape libraries filled with tapes that physics and time have effectively erased. Instead of thinking about keeping a stash of outdated tape drives around to run tapes in an outdated format “in case you have to go back,” think about migrating valuable data to virtual tape libraries or other technologies that can take you forward.
Of all the rooms in your house, probably the most stressful one to confront in preparation for moving day is your attic. Everything in it has some unfinished story attached to it. Typically, it’s where all the postponed decisions hide. You may not have a designated attic in your data center but every data center has metaphorical attic full of stuff stored under “we’ll-deal-with-that-later” label.
You don’t want to drag all of those unfinished decisions and irrelevant what-ifs into your new environment. Ironically, despite all the effort that goes into putting off confronting them, taking the time to actually make all those unfinished decisions can be very liberating.
A new beginning
The converged infrastructure in your future is as much a new mindset as it is a data center full of new technology. It’s not enough to just clear out the miscellany lying around your existing data center. Before you migrate to your new converged infrastructure, you need to establish a shift in attitude appropriate to your new surroundings.
Instead of the reactionary tendency to keep everything “just in case” that was reinforced by the rigid, silo environment you are leaving behind, begin thinking proactively about leveraging pools of resources on demand in a dynamic, flexible, integrated environment. As you begin to spend less time on provisioning and physical resource management, you will have more time to devote to bringing true value to your organization: better alignment with your core business objectives, proactive management of the environment, and improved IT processes that give you the time and freedom to take projects to full completion — including decommissioning and removal of the trash.
A converged infrastructure is all about giving you the ability to dynamically respond to change. To take the most it, when you move into your new data center, you need to leave your fear of change behind.
Check this space for the last article in this series: Don’t Have a Party Just Yet when Logicalis technologist Brandon Harris provides some guidelines for evaluating, modifying and developing new processes and procedures for working in a converged infrastructure environment.
Other articles in the series include:
Why Can’t We all Just Get Along? (Part 1/6)
How to Sell the CEO on Change (Part 2/6)
It Always Comes Down to People (Part 3/6)
Don’t Let Your Legacy Become Your Legacy (Part 4/6)
Brett Anderson is the director of HP Server & Network Solutions at Logicalis, an international provider of integrated information and communications technology solutions and services, where he is responsible for the support and growth of the HP Networking and Server business within Logicalis. Mr. Anderson has been with Logicalis fourteen years, with consulting and management responsibilities in Logicalis’ Professional Services and Sales organizations.