Making the Case for OpenStack: Critical Success Factors (Part 2)
Last week I wrote about some of the challenges to successfully implementing OpenStack in the enterprise. The biggest obstacles have nothing to do with technology, but rather have to do with Governance, Processes and Skills.
Related to this topic are the Critical Success Factors, or CSFs, of your OpenStack strategy. CSFs make or break a strategy, they drive and guide an organization towards success, especially when it comes to Information Technology implementations that require cultural change.
There is always skepticism when I introduce the concept to clients, too “management consulting”… but the reality of the matter is that if you don’t apply a relentless focus on the CSFs, your implementation will most probably fail.
Here are the more common CSFs to consider when executing your OpenStack strategy:
Executive Sponsorship – Introducing such a revolutionary concept into an enterprise will only succeed if you have the blessing and support of senior executives. Not everything will go the way you expect, there will be setbacks, some larger than others, and you will need Executive support in order to overcome the challenges you will face. They force you to stay committed to the journey.
Governance – Cloud is a catalyst for breaking down cultural barriers and shattering existing fiefdoms in an organization. If you think you can bulldoze your way through years and years of legacy processes and culture, think again. You must create a Governance structure that is inclusive of all relevant lines of business, ensure that channels of communication are always open, and that the objectives are clear from the very beginning, with all participants on board to execute the common vision.
Position for Services – As you migrate to a cloud environment, where there is more automation and self-service expectation, your IT, Engineering and Operational organizations will need to shift from a linear, waterfall project-based approach to one of shared services. This is easier said than done, but it can be. Introducing DevOps processes, agile methodologies, and common shared services incrementally will enable the shift to a truly agile enterprise.
MVP– As with any new technology, one that will disrupt the status quo, you will need to show successes early and often. Don’t try to boil the ocean, don’t architect and build a massive cloud right out of the gates, start small, with a PoC, transition to a Pilot, run “safe” production workloads, be willing to fail in order to make the solution more resilient in the long term. A Minimum Viable Product (MVP) approach is the right way to go. There is no big bang with cloud.
KPIs – I can’t stress this point enough, you need to clearly articulate the success metrics and measure early and often, to demonstrate success but also to continuously improve the processes and platforms. Not having a reference point will make it difficult for you to justify the investment that you will need to make, over and over, to senior executives. Categories typically include financial e.g. TCO, cost avoidance, IT efficiency ratio; agility e.g. cycle time, provisioning time, MTTR; and quality e.g. infrastructure availability, defect backlog.
Bottom line, don’t neglect the softer aspects of your OpenStack implementation. If you want to take OpenStack mainstream, it’s time to stop treating it like a science project and apply the same discipline and approach as you would to any strategic initiative.
Author: Francesco Paola