External Private Clouds: How They Should Work
How should external private clouds operate, and what services should they provide.
In our last look at the topic, we defined what external private clouds are along with some of their benefits and possible downsides. To recap, a “cloud” is typically a compute model based on aggregation of standardized infrastructure in which compute resources have been pooled and managed as an abstraction, typically through virtualization, according to John Sloan, lead analyst at Info-Tech Research Group. In a cloud services model, the tenant/client/customer pays for only the resources used, and resources can be easily scaled up or down to meet demand.
With external public clouds, the infrastructure is owned and managed by a third party, and numerous renters of capacity coexist in a multitenant environment. In the external private cloud, the provider establishes a dedicated block of capacity and resources that are not shared with other third parties.
Beyond the definition, though, how should external private clouds operate, and what services should they provide? Without a clear answer, establishing the value of an external private cloud becomes difficult at best.
According to David Kramer, senior vice president of product and client services for External IT, the three central non-infrastructure elements of a virtual private cloud (his company’s term for external private cloud) are automation, a strong service department, and a stack of services covering the cloud comprised of support, administration, and security/compliance. This services stack will be comprised of both automated systems as well as people. All of these factors will serve as enablers for the top function of any effective private cloud: a unified, universally available, elastic service platform.
“You might have three hundred users,” says Kramer. “Half of those work in the main office, but only 75 are ever in the building at once. The other 150 are scattered around Europe and Asia. But everyone needs access to the same applications and data in real-time on any device. You don’t want the mobile workers stuck with Google Mail while all of the office workers use a best-of-breed desktop solution. You want everybody going in through one centrally administered portal that works on any and every device the workers are using, and that portal platform needs to support all of the businesses applications, not just one or two.”
Elasticity means that all of this universality can scale both up and down as business needs dictate. Cloud providers are excellent at scaling up—this is one of the great advantages of the cloud model. But what happens when business contracts? Customers shouldn’t be penalized for downsizing. By the same token, while some providers still offer flat rates on set contract periods, a better option is usually to be charged based on usage, like a utility model. This addresses the issue of only having 75 out of 150 employees present at any given time. You don’t want to pay for 150 seats around the clock; you want to pay for the actual compute resources consumed.
Universality of device support is often a deal-breaker with external private cloud providers. Not only must the mobile OS platforms be supported, but the requisite applications must be supported, as well. That includes cross-platform, centralized management. Some providers will advertise cross-platform compatibility, but this doesn’t assure that every app for every platform can be efficiently administered, logged, and maintained. Ask for proof before trusting assurances on this point.
Also tied to Kramer’s three central elements is the issue of setup.
“The way that they set themselves up to on board large quantities of things is someone has to go create a spreadsheet, get a bunch of information lined up in the spreadsheet, upload that, and then you get the base templates. Then you go person by person, taking the unique things for each one, and you get to clean them up. My portal technology allows me to onboard, say, 1000 users with all their mail set up, all their SaaS apps set up, all their files, printers, and so on. Everything that’s in their IT environment gets turned on and enabled at the time those 1000 users are ingested into the portal.”
Other considerations for a robust external private cloud should include disaster recovery(DR). Natural disasters destroying data centers are far less common than human error causing a runaway chain of mishaps that wipe out a database. Enterprises should be running a DR solution anyway; it may prove more cost effective and easier to implement if integrated into a broader external private cloud package.
Not least of all, if many apps are involved in the cloud deployment, managers should be asking about having single sign-on tools to help streamline access and reduce support calls.
William Van Winkle has been a full-time tech writer and author since 1998. He specializes in a wide range of coverage areas, including unified communications, virtualization, Cloud Computing, storage solutions and more. William lives in Hillsboro, Oregon with his wife and 2.4 kids, and—when not scrambling to meet article deadlines—he enjoys reading, travel, and writing fiction. See here for all of William's Tom's IT Pro articles.
(Shutterstock cover image credit: Cloud Computing)