Matching the Cloud to the Workload

Arthur Cole
Slide Show

Multi-Cloud 101: 7 Things You Need to Know

Cloud providers want enterprise workloads, and the enterprise wants to push more data and applications to the cloud. Sounds like a perfect match, doesn’t it?

Well, yes and no. While it is true that enterprise-class cloud deployments are expanding at a steady clip, and more of these are taking on real production workloads rather than bulk storage and data backup, many organizations are still struggling with the generic nature of cloud resources.

For decades, the enterprise has had the luxury of crafting highly customized infrastructure whenever it was necessary. It was one of the perks of building and maintaining your own data environment. This is certainly possible in the cloud, of course, but it often comes at a higher cost, since the economies of scale are not the same as with generic workloads. Where there is a need in business, however, there is usually someone willing to fulfill it, and the cloud industry is rapidly transitioning from a basic level of functionality that caters to consumer tastes toward the more specialized requirements of the enterprise.


A case in point is Infor, says Diginomica’s Derek du Preez. Once mired in on-premises enterprise apps, the company has transformed itself into a provider of customized cloud environments targeting a wide range of vertical industries. Infor provides integrated cloud suites that allow clients like Whole Foods and Travis Perkins to leverage customized environments on AWS and other clouds. In this way, organizations gain immediate support for emerging initiatives like data mobility and social collaboration while at the same time adhering to Infor’s own ION and Hook & Loop Soho architectures. Infor officials say that its cloud business has grown from 5 percent of its revenues to more than 50 percent in the last three years, with more than 65 million users across 7,000 enterprise clients.

Other firms are targeting the cloud as a means to support emerging technologies that most legacy data center environments cannot handle without major upgrades. Machine Learning (ML) is one of the leading contenders, says Datanami’s Alex Woodie, as it is already a common facet of Big Data infrastructure for finance, manufacturing and other IoT-leaning industries. This is spurring specialty software shops like Connecticut’s Cycle Computing to create new tools aimed at simplifying the movement of ML workloads across hybrid clouds, which more often than not requires the deployment of specialized cloud environments to gain the proper RAM-to-CPU ratios to support intelligent, data-drive applications.

Efforts are also underway to isolate key workloads in the cloud to enable higher levels of performance and security. Bracket Computing recently launched a new component to its Cloud Workload Protection Platform called the Computing Cell that enables specialized policy controls to accompany workloads as they transcend public and private infrastructure without diminishing the performance of self-service cloud environments. The system is built around the company’s Metavisor, a virtualization layer that sits between the guest OS and the cloud hypervisor and inserts security and other services into the data path so resources can be optimized for a given workload. In this way, the enterprise can tailor the cloud to its liking while maintaining high visibility into the workload plus an added layer of security and encryption.

The cloud has long prized convenience over optimization, effectively doing an end-run around IT by providing basic services to line-of-business managers much more quickly than traditional infrastructure, and with a lot less complexity. But now that higher-order workloads are being targeted, it turns out that a little complexity is warranted, as long as the end results are solid.

The cloud has greater propensity for change than most legacy environments, so there is no reason to think it can’t handle the important work, but it will require careful coordination of many moving parts on the physical, virtual and application layers.

And it will require broad recognition by cloud providers and users that cheaper and easier are not always better as the workloads become more critical.

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.

Save



Add Comment      Leave a comment on this blog post

Post a comment

 

 

 

 


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

 

null
null

 

Subscribe to our Newsletters

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