Attack the Stack: 5 Tips for Successful Cloud Deployment
The enterprise is starting to look at the cloud as a data environment in its own right, not as just a low-cost adjunct to the data center. But in making this transition, organizations must confront a number of thorny issues regarding what to put on the cloud, how to do it, and what type of cloud is warranted.
But even as this change of perception takes root, planners are already working to pull the rug out from under our feet by recasting the cloud from a collection of efficient and flexible resources to groups of applications and services.
At the moment, says SADA Systems CEO Tony Safoian, cloud deployments are being driven by several factors. First, in a reversal from earlier impressions, organizations are starting to realize that data is in fact more secure in the cloud than at home, removing the last great barrier to widespread deployment. And with Big Data and IoT workloads coming down the pike, enterprises are eager to tap into machine learning, containers and advanced mobile technologies, which can be done much quicker and at less cost in the cloud than by building out data center infrastructure.
Not everything is suitable for the cloud, however. According to SolarWinds’ Gerardo Dada, there are a number of key criteria when it comes to determining what should and should not leave the data center. Applications with multiple dependencies, such as CRM and ERP, might have trouble in the cloud, while a self-contained company blog would not. As well, web-based customer-facing apps will probably perform better in the cloud, as will processes that involve varied workloads that rapidly scale resource consumption up and down. Overall, organizations should only deploy onto the cloud if there is a tangible benefit such as cost reduction or performance enhancement. As the old saying goes, if it ain’t broke, don’t fix it.
To providers like Amazon, however, everything in the data center is already broken, so enterprise infrastructure will become all-cloud simply as a matter of course. In fact, the very existence of the so-called “serverless” architecture Lambda is intended to get the enterprise to stop thinking of the cloud as infrastructure – full of virtual machines, containers and other techy things – and start thinking of it as processes and functions. The company recently bolstered its cloud with the Kinesis Analytics platform, which provides a service-based approach to handling streaming SQL workloads by automating mundane tasks like provisioning, deploying and scaling resources. In this way, even people who don’t know the first thing about programming or resource management can build the applications they need and launch them into production environments.
This is what “cloud-native applications” will actually look like, says InfoObjects CEO Rishi Yadav. Rather than simply offloading traditional apps into IaaS deployments, or tweaking them a bit for either PaaS or SaaS support, a serverless environment provides end-to-end lifecycle support for advanced functions, from development to retirement. At the same time, it allows multiple functions to be mixed and matched, perhaps automatically or even autonomously, to create new services that could not even exist on traditional data infrastructure.
Not having to deal with infrastructure, resources and all the integration/optimization hassles that go with them is undoubtedly mighty appealing to many enterprise executives. But like a standard-shift transmission, there is something to be said for maintaining some control over what you are doing and where you are going.
Complete hand-over of all things IT to the cloud may be in the cards for many organizations, particularly start-ups looking to upend established business models, but it isn’t likely to become universal any time soon. Productivity is still a driving force in business, and outsourcing the ability to control productivity will be a tough sell for the cloud.
Arthur Cole writes about infrastructure for IT Business Edge. Cole has been covering the high-tech media and computing industries for more than 20 years, having served as editor of TV Technology, Video Technology News, Internet News and Multimedia Weekly. His contributions have appeared in Communications Today and Enterprise Networking Planet and as web content for numerous high-tech clients like TwinStrata and Carpathia. Follow Art on Twitter @acole602.
