Question: How can cloud migrations cut through the mess of legacy applications and failed IT initiatives?
The growing IT best practices consensus is that creating enterprise clouds are smart money for corporations. Private cloud infrastructures empower user with self-service portals, stable application building platforms, and flexible service delivery models. In addition, warming the hearts of corporate bean counters around the world, enterprise clouds can save significant amounts of money by maximizing utilization rates of expensive data center systems. Cloud nirvana is achievable; there are plenty of successful case studies around proving the value. However, for many senior IT executives just a review of the corporate IT portfolio is a daunting task, let alone systematically determining which applications should move to the cloud, then figuring out the best way to achieve that goal.
The sorry reality is that no matter how well disciplined a company is about its IT governance, the average enterprise has 20 plus years of legacy applications in its portfolio that it needs to manage. Over time, even the best run company will acquire what one of my co-workers affectionately refers to as, glops and legacy linguini. We all know the recipe; start with a couple of semi-successful enterprise application implementations (the kind where the company is using both the old and the new systems and cannot shut down either), throw in a few mergers and acquisitions (include a handful of management regime changes and reorganizations just to spice things up a bit), then stir in thousands of undocumented patches, temporary fixes that turn permanent and shadow IT applications. Bake until solidified into an impenetrable morass of siloed functions and tangled application inter-dependencies.
Fortunately, there are some emerging tools, methodologies and best practices that can cut through layers of ossified applications, and offer real actionable guidance on the right approach to moving an enterprise application portfolio to the cloud. If you are seriously considering adding a private cloud to your IT portfolio, here are just a few questions to ask about your portfolio to get started:
- How ready are the applications for migration to the cloud? – There are tools available that can drill down within each application to determine the exact amount of effort it will take to make it cloud ready.
- Would moving the applications be operationally disruptive? – Again, an organizational cloud readiness skills assessment provides guidance to how the IT organization needs to change to meet the different requirements of managing an enterprise cloud.
- Are they built on x86 based Windows or Linux platforms or are they Solaris, HP-UX or some other more exotic platform? — Only a few years ago I ran into an ERP system that was still running on a Data General platform. A rewrite of the application was a foregone conclusion in that case.
- How much of the existing portfolio is already virtualized? – Virtualization is not cloud (a common mistake), but being virtualized makes moving to the cloud that much easier. The average enterprise is typically only 30% virtualized.
- Can the application be replaced by existing SaaS applications that offer similar functionality at a fraction of the cost? — Cloud email, document management and supply chain applications for all except the most unique circumstances are far superior to the legacy systems they are replacing.
From the inside undertaking an enterprise cloud project might seem hopeless, but these types of situations are where bringing in outside experts with real expertise in cloud infrastructures and application transformations can really pay off. I can tell you from hard earned personal experience, we have seen it before and whatever mess your portfolio is in, others were far worse and they successfully moved their portfolios to the cloud.
About the Author
Beth Cohen, Cloud Technology Partners, Inc. Moving companies’ IT services into the cloud the right way, the first time!