Building a More Automated Cloud

Arthur Cole

If 2012 was the year the enterprise began building cloud infrastructure in earnest, then 2013 will likely focus on making them work.

However, there are many ways to view a successful cloud deployment. As I mentioned yesterday, the cloud represents much more than simply a means to provide the same old IT at a lower cost. Dynamic data flexibility, virtual unlimited scalability, advanced social networking and a range of new possibilities are all in the offing, but that means the cloud will not merely have to function, but function well.

To that end, the cloud will need quite a bit of automation — more so than current virtual environments. As enterprises seek to unify their internal and external infrastructures, most likely through a hybrid environment, the need to automate data and resource interactions will become paramount. According to Justin Crotty, senior vice president and GM of cloud service provider NetEnrich, expect to see a raft of integrated solutions hitting the channel soon as software developers, systems integrators and other players look to unify infrastructure, application and workload automation in a bid to increase the value of cloud-based offerings.

This cuts to the heart of what many see as the chief stumbling block in the cloud — the need to coordinate multiple applications and services across disparate infrastructure. As BMC Software found out in a recent survey, more than 90 percent of enterprises report critical job failures, file losses and other major disruptions due to incompatible automation systems. The company recommends a shift toward workload automation as a means to bridge tasks like job scheduling across hardware platforms, applications and physical location.

Of course, cloud infrastructure itself has proven to be highly dynamic, leading to a need to automate the process of building up and tearing down the various elements that application environments come to rely on. To keep track of all this, companies like Opscode have devised infrastructure automation platforms like the Private Chef and Hosted Chef systems to both speed up the configuration process and keep track of all the changes. The latest iterations include updates to the configuration management engine to carry policies across multiple resources, plus integration into leading software development lifecycle (SDLC) platforms to rein in build, test, configuration and deployment functions.

Automation is always easier if you control facets of the cloud stack, like Oracle does. The company recently upgraded its Exalogic Elastic Cloud with new server-level virtualization capabilities that allow users to build IaaS environments without having to configure physical server, storage or network resources. At the same time, the platform features new systems orchestration tools that oversee metadata related to various environments so they can be repeatedly and consistently deployed across the cloud. The drawback is that the system has been optimized for Oracle business applications, making it difficult to introduce third-party ERP and CRM software.

Like the virtual layer it rests on, the cloud is proving too complex for mere mortals to manage. As data, applications and even infrastructure come and go at the drop of a hat, only an advanced automation stack will be able to keep track of it all.

This will require a level of trust that many IT managers will have trouble providing, but it is crucial that rote automation tasks be handed over to software in order to capitalize on the cloud's unique data capabilities.

And ultimately, the challenge for humans will be to establish the policies that govern the automated environment, not the automation itself.

Add Comment      Leave a comment on this blog post

Post a comment





(Maximum characters: 1200). You have 1200 characters left.




Subscribe to our Newsletters

Sign up now and get the best business technology insights direct to your inbox.