IT professionals have the basic ingredients to cook up their own cloud-like infrastructure-but there's no recipe, and many ingredients don't combine well. Especially as infrastructure innovation evolves in a way that doesn't necessarily deliver better efficiency and instead, drives up complexity. Yet, the most common answer infrastructure and operations professionals recently told Forrester that what motivates server virtualization adoption was the need for greater efficiency and sharing of the IT infrastructure. Over the years, vendors have tried to accommodate this need, but have often provided solutions that left too much of the burden on their customers-until now. Recent integrated solutions take a big step toward delivering complete virtual infrastructures that just might solve some of the major issues IT professional face.
Past vendor solutions were factory-integrated and built, tested and supported as a single solution to solve a specific purpose. However, their highly proprietary nature made them closed solutions with narrow use cases that left organizations at the mercy of one vendor's road map. As a result, these solutions met with only moderate success.
Chapter two of this has now begun, and its prospects are much greater, as these solutions are focused on helping customers create complete virtual infrastructures. And, rather than delivering proprietary solutions, key vendors, such as HP, IBM, Cisco and Dell, are delivering prepackaged server, storage, network and management solutions that use a preselected best-of-breed approach.
With many of the same benefits of past solutions, these new options, such as Dell's Virtual Integrated System or IBM's CloudBurst, let customers take the next step with virtualization by providing a unified approach that integrates management tools and element managers to complete requests through the entire system. Additionally, an easy transformation from a virtualized infrastructure to an internal cloud becomes feasible as most of the core elements of a cloud infrastructure are present. Tools such as automation solutions for deployment, management and monitoring, and the self-service portal for deployment requests are often included.
But, just because there is now a purchasable interpretation of what an internal cloud infrastructure should look like, this doesn't mean it's the right interpretation for every organization. When considering a packaged solution, it's essential to evaluate whether the vendor's interpretation maps to the operational processes of the organization, and, most importantly, if the organization is even ready to accept the solution as delivered. Consider these four key assessment points before moving forward:
1. Assess your virtualization maturity. If your group isn't experientially ready to standardize how workloads are deployed and doesn't trust automation to a great extent, you probably aren't ready for these types of converged infrastructures.
2.Start down the path with test lab resources. Accelerate learning on how to operate a converged infrastructure by proving it with noncritical functions. This is a great way to relieve infrastructure teams from the constantly changing burdens of lab setup, teardown, and maintenance.
3.Resist trying to outsmart the vendors by breaking this apart. One of the biggest benefits of pre-integrated bundles is the reduction in variability of configuration they bring to the table. So while it may be enticing to customize to substitute known products, it will only dilute the value and make it harder to support down the road. Your time is better spent on innovating business technology solutions, instead of troubleshooting individual technology components.
4.Consider having someone else manage it for you. If the organization isn't ready to administer a unified infrastructure, but you want to bring the value to the business now, it is certainly appropriate to ask a managed services provider to operate these solutions on the organization's behalf.
Because of the above stipulations, one option infrastructure and operations professionals can consider is to simply view the unified solutions as reference implementations or blueprints for the future, rather than as practical solutions for the present. Or, on the opposite end of the spectrum, start anew and accept the best practices in these implementations, leaving the evolution of the existing infrastructure for slower transformation. Forrester spoke with several customers who are doing just that, leveraging unified packages to build fresh environments for new workloads coming from the business. Then, over time, older systems will be integrated with the fresh-start infrastructure as a means for transformation.
Either way, as converged infrastructures make their way to the market, the question arises: do these solutions pose a threat to the infrastructure architect? If the architect's value is based on the ability to make disparate pieces fit together on a custom basis, then maybe so. But if value is determined by the ability to architect processes and solutions that enable the business to grow faster, more efficiently, and more flexibly-there's still plenty of work to be done toward this end.
James Staten is a Principal Analyst at Forrester Research, where he serves Infrastructure & Operations professionals. His research provides insights into server-centric issues including infrastructure consolidation and virtualization, data center migration and reconfiguration, cloud computing, and high performance and technical computing infrastructures.
Read more about cloud computing in CIO's Cloud Computing Drilldown.
Join the CIO Australia group on LinkedIn. The group is open to CIOs, IT Directors, COOs, CTOs and senior IT managers.