We’re beginning to move along in the cloud conversation. More organizations are seeing the benefits of moving towards some type of cloud infrastructure. Still, even though use-cases are being developed for a cloud migration, there are still some concerns around execution and finding the right cloud model. With the evolution of the cloud platform, it's a good time to look at a few considerations when working with the modern cloud.
Remember, there are a lot of new technologies spanning the cloud platform. We now have various “as-a-service” models and even more virtual and cloud workloads being delivered.
In working with cloud computing, organizations can quickly see how they’ll benefit from such a powerful platform. However, four major considerations must still be addressed by many data center and IT 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:
- Number of uses connecting – concurrently.
- Number of applications, desktops or workloads residing in the cloud.
- The amount of traffic to be pushed through the WAN – whether private or public.
- Distance that the data has to travel. WAN optimization may have to be an option.
An organization can have the best technologies deployed at the data center level. However, if the proper bandwidth and WAN considerations aren’t applied, the user experience will suffer.
- 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.
Working With the Right Cloud Model
There are now four distinct cloud models to choose from. Each has a specific use-case and will benefit an environment differently. At a high-level, these are the four major cloud environments to work with:
- 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.
In moving to a cloud solution, understand the options and see the ROI. With improvements in API structures, cloud platforms (and data center infrastructure) as well as the ability to seamlessly integrate cloud services – it’s time you explore the power of the cloud.