Clouds Without Fear: The Big Bang Theory of Cloud Sprawl

Cloud Sprawl is fast becoming the new enterprise reality, but it does not need to be something to be feared. Rather than avoiding cloud sprawl, enterprises can learn how to identify cloud sprawl, rein it in and make it work for them by leveraging tool chains now available for managing workloads spread across multiple clouds. To do this successfully, you need to understand the Big Bang Theory of Cloud Sprawl

The Big Bang Theory of Cloud Sprawl

The rise of cloud sprawl in the IT world is the result of a collision of a forces at work in today’s enterprise, a phenomenon that I’ve labeled “the Big Bang Theory of Cloud Sprawl.” Cloud sprawl is actually the result of the by product of the diaspora of applications, easy access to cloud resources, and corporate cost-cutting.

Reining in Application Diaspora

If enterprises had just a dozen applications to manage, life would be easy for IT departments. Unfortunately, the modern enterprise is faced with a diaspora of applications. Mobile applications of every kind find their way into the hearts and iphones of the corporate fleet while numerous custom applications run continuously to meet specialized business needs. Some of these apps are zombies and really have no place in the enterprise. Yet they are kept alive. In the past, enterprise applications ran on mainframes or server farms. With the advent of cloud technologies, these apps are fast migrating to the cloud, creating sprawl.

Collision of Easy Access and Corporate Cost-Cutting Mandates

The application deployment paradigm has turned on its head with the rise of cloud technologies being made available “as a service.” The easy access to public cloud computing resources facilitates the resurgence of shadow IT projects. As developers increasingly use services such as AWS, Rackspace, Heroku, or Engine Yard, they create an infrastructure diaspora.

While running around trying to put out these fires and rein them back into compliance, IT teams are pushed to migrate core services and applications onto one of the many flavors of cloud infrastructure that have emerged in order to take advantage of the cost benefits. These include AWS, OpenStack, CloudStack, VMware’s vCloud, and KVM.

Solving the Cloud Sprawl Equation: It’s All About Management

It probably sounds very contrarian, but a properly managed cloud sprawl is a good thing and should be viewed as a sign of a healthy enterprise cloud strategy. With good cloud management and configuration tools, cloud sprawl should not be viewed a problem, but rather the outcome of a well-run IT cloud cost management program that takes advantage of low cost cloud computing resources without risking security, compliance or effectiveness.

With a well-managed cloud sprawl, organizations recognize that different clouds serve different purposes. Low cost public clouds and secure private clouds can live side by side in the enterprise. There are times when low-cost public clouds best serve the needs of an organization and times when a private cloud is the right choice to enforce compliance. The option to run applications on any cloud is the best policy to ensure redundancy and fail-over. The Big Bang Theory is about harnessing cloud computing to the enterprise’s advantage.

Smart Management with Private PaaS

Solving the cloud sprawl equation takes a few steps. First, IT can easily manage the application diaspora with a polyglot private Platform as a Service (PaaS). Applications depend on a multitude of frameworks, services, web servers, and databases that defy ordinary standardization practices. A polyglot-friendly PaaS architecture is the only way to maintain control and oversight. Private PaaS ensures that all libraries, languages, frameworks, and other stack components are staged, deployed, and controlled in the cloud from a single pane of glass. IT can enforce business rules, administer cloud resource allocation, track usage, uncover zombies, and determine where application usage bottlenecks are occurring.

Second, private PaaS gives developers the power of code to cloud deployment, easy access to computing resources and to new technologies including buildpacks, services, databases, web frameworks, and git hooks. In turn the C-level management team obtains access to the latest technology that ensures compliance and competitive advantage.

Third, it is important to keep in mind cloud portability. A PaaS that can run on any cloud will allow the enterprise to take advantage of the best offerings from cloud hosting infrastructure vendors as they compete each other with lower cost and higher volumes. With a cloud infrastructure-agnostic private PaaS, applications can run on any cloud that an enterprise chooses according to specific requirements in terms of cost, availability, security, regulatory compliance, or redundancy.

Finally, it is all about management. With private PaaS, IT gains complete oversight of applications through an advanced managed console, system level monitoring tools such as nagios and zenoss, application monitoring tools such as loggly, papertrail, and newrelic, and multiple PaaS cluster management tools such as Appsecute and CloudAbility.

The right management tools & techniques enterprises can turn cloud sprawl into a competitive cost-cutting advantage.

About Diane Mueller

Diane Mueller is a bit of a PaaS Queen & the Cloud Ecosystem Evangelist for Red Hat's OpenShift Origin the Open Source Private PaaS. She is a leading advocate for Open Cloud technologies, has authored numerous articles & white papers on emerging technologies and can often be found pontificating on various and sundry dev & open source topics at conferences ranging from OSCON, PyCon to CloudExpo, CloudConnect. You can follow her on twitter at @pythondj