If you are going to change your solution, why not change all of your solutions?

I was reading a blog by a coworker a month ago, and he said “If you aren’t going to change your process, why change your solution?”, or something to that effect.  It stuck with me, obviously.  It was a great blog on IT transformation from a people/process perspective.  I often say “technology is the easy part” when discussing IT transformation, but I’d like to retract that.  Implementing a new piece of technology is the easy part.  Displacing a set of existing solutions to build the new nirvana is not as easily achieved.  In fact, it is so hard, that most companies ignore it.  Forever.

So many sigh with relief when the words “green field” are used as the basis of an upcoming project.  And most everyone cringes at the idea of integrating with and upgrading the “brown field”.  There are business critical applications responsible for thousands of dollars in revenue daily that often function magically running in this black box. Unraveling the rat’s nest could mean dire consequences for the company and could require updating your resume if all is unsuccessful.  However, we cannot continue to look the other way and hope someone else addresses this issue.  Unlike the first days of Active Directory and Exchange, where it was easier to stand up a new system and migrate everything over, now we have thousands of servers and applications to address.

Regardless of the complexity or sheer magnitude of this undertaking, it is imperative that we solve this.  If not, we are only digging ourselves into a deeper hole.  But, the business and IT benefits that justify moving forward with our exciting projects will be canceled out by the costs of managing and maintaining the untouched environment if it remains.

I was recently working with a customer who has a great private cloud implementation, one that ranks in the top 5% of true private clouds in existence.  We were strategizing about the three year roadmap as it relates to business initiatives.  One of the metrics was mentioned was that the CIO wishes to reduce OpEx spend in IT by X amount.  Improving their private cloud would be one place to reduce OpEx by increasing automation, etc.  However, we discovered they still have thousands of physical x86 servers in locations other than the main, strategic datacenters.

We could easily meet the CIO’s reduction in OpEx goal by performing a large datacenter migration while virtualizing those systems rather than trying to squeeze out more efficiencies from their current private cloud build.  It won’t be new and fun, but it is the path with the greatest return.

You might be thinking, I am 95% virtualized, P2V is so 2009.  Great!  You can consider the next iteration in this process when most systems have been virtualized and the organization is moving down the IAAS and PAAS path.  It is likely the management and monitoring tools for these systems are actually from 2009.  Also, it stands to reason these tools do not understand nor function with cloud constructs.  It is very common that these ancient tools are only used for 5% of their, albeit questionable, abilities.

Rationalization of management and monitoring tools is a phenomenal way to reduce CapEx and OpEx spend.  The yearly spend on that software’s maintenance bills and footprint expansion will be avoided.   A rationalization project might not be met with enthusiasm like the deployment of automated day 2 operations.  However, it is the right thing to do.  And it will make the lives of everyone managing the cloud far easier.  This approach will ensure the success of the new platform because the new cloud management system will be able to understand the concepts and flexibility that is cloud.

I will summarize with the title, “If you are going to change your solution, why not change all of your solutions?”  If we only cherry pick the exciting, new technology projects and limit ourselves to deploying a portion of these architectures, we are not meeting the end goal of this new IT.  We want to provide choice, agility, faster time to market, and a reduced cost for everyone, not just those with brand new cloud-ready applications.  Welcome to operating in 2015!