4 Key Cloud Considerations For Data Center Managers

Data migration and control. Initially, this was a major concern for organizations. Where is my data being stored? How is it being accessed? Do I even have control over it? Now, better technologies which are directly in-line with the cloud mentality are easing these concerns. Consider this: new data loss prevention (DLP) and onsite cloud options are providing administrators with granular control over their data and how it’s being distributed. Furthermore, storage systems are becoming more “virtualization aware” and cloud ready. This means that some enterprise SAN systems are now able to live migrate data from one controller or disk aggregate to another. Now, imaging being able to do that from one cloud platform to another. Powerful stuff.

WAN, Cloud and Bandwidth. Cloud computing is, at a high-level, just the transmission of data over the Wide Area Network or the Internet. Sometimes organizations use their own private cloud platform, but for the most part, they still have considerations around bandwidth and WAN utilization. One of the biggest challenges around the cloud is the understanding of how the WAN and bandwidth plays a role. When designing a cloud solution, remember the following:

Application compatibility in the cloud. That’s the beauty of the cloud. For the end-user there is complete transparency. However, at the data center level, there is complete control. Cloud computing works very closely with virtualization. And, because of that, creates a simpler platform for application compatibility. For example, you’re able to run a 32-bit application as well as a 64-bit application in the same environment while still presenting both apps under the same portal to the user. Application virtualization in conjunction with cloud computing has come a long way. Products like Citrix XenApp create a platform where heterogeneous applications can live side-by-side peacefully while still being presented in a single interface to the cloud user.

Private – If your organization needs to stay locked down, but has remote users who need to access a central data center for applications, desktops, or files. This is the right model for you. In this scenario, you are deploying your own hardware and software to facilitate the cloud solution.

Public – Many times organizations need to test out an application or a set of servers. Instead of buying equipment, they’ll “rent” it out. In this pay-as-you-go model, IT managers can provision VMs in a public cloud environment, work with them as long as needed, and then decommission those servers. All of this will have little to no impact on the local corporate data center. Hybrid – In some cases, two cloud platforms need to be combined to test production data with a public cloud VM. In this case, organizations are able to create secure connection between the public and private environment – effectively creating a hybrid cloud.

Community – This is a newer and most recent addition to the cloud model family. Imagine that there is an application being hosted by a provider. Instead of providing the same application for a customer on individual VMs. The provider is able to logically segment connections coming into its environment. From there, it will allow multiple customers to connect into its cloud platform to access that single application. This is, effectively, a community cloud.

As with any newer technology, don’t let the concept overwhelm you. The flexibility of cloud computing offers the flexibility in the deployment option. By far, the most important part of any cloud (or IT) initiative will be the planning phase. When it comes to cloud computing, think short and long term. The power of an elastic cloud can allow an organization to scale as business needs dictate. However, a poorly implemented cloud solution can become a fast budget drainer.