IT Transformations Don’t Have to be Scary

The word transformation typically implies some daunting task, especially when applied to large IT Infrastructures such as data centers and mission critical application platforms.  The truth, however, is that transforming a data center doesn’t have to follow a “Big Bang” approach.  It is perfectly feasible to transform small pieces of your infrastructure incrementally.

When building architectures for an IT environment, it is typical to take one of two common approaches.  The first is to create a high-level solution that solves numerous problems, but is either too complex or takes too long to implement and, consequently, is shelved and never looked at again.  The other approach is that oftentimes solutions are engineered specifically to the problem at hand, without considering the impact on the remainder of the infrastructure or long-term complications.  Neither of these approaches ends up being transformational.

A successful transformation project avoids the problems that accompany the aforementioned approaches by taking the best pieces from both, and avoiding the pitfalls.  First of all, an end-state must be defined, but need not necessarily be set in stone.  It should be based on standards and needs and adaptable as new problems arise so it doesn’t become shelfware.  Secondly, the focus to transformation should be geared toward solving existing problems—while keeping the target architecture in mind.  Problems should be solved with capabilities that fit into the context of the target architecture, and most importantly can be reused.  In this manner, a solution is being engineered that isn’t monolithic in nature.

This approach to transformation becomes much easier to manage if each solution is viewed as a capability or a combination of capabilities.  For example, as part of the transformation effort, if an application is running into memory-related data-access constraints, the solutions that could help resolve the problem could include a distributed data caching capability, a memory appliance capability, or both.  Since we want to solve problems with the target architecture in mind, it is important that the solutions be reusable.  Thus, if another application runs into similar constraints, it can simply leverage the existing solution instead of having to go through yet another design step or process.

While the above example is very simple and straightforward, it helps illustrate an important concept about capabilities that can be applied to even the most complex problems.  Once a core set of capabilities have been introduced into the environment, they can be expanded and reused over and over.  This makes transformation happen organically, primarily by reusing components, and introduces new capabilities where necessary.  All the while, business problems are being solved on a regular basis, and the transformation projects retain credibility by repeatedly demonstrating success.

Tags: Featured

Category: Datacenter Transformation, IT Management