The modern IT infrastructure has adopted a new type of technology which has helped it become more agile and efficient. Virtualization has now come far beyond the server. In many environments, IT administrators are considering application, security, storage, and even desktop virtualization platforms. Amidst all of the various options an organization can go with – virtualization and cloud services delivery – has been something many IT departments are looking to consider deploying. The idea is simple, empower the end-user and deliver a powerful virtual experience to any device, anytime and anywhere.
On the outside, executives can immediately see the benefits of working with this type of technology. However, cloud computing and virtualization are not just one technology. To make it all work effectively, there has to be an underlying infrastructure to support such a platform. In some cases, cloud computing can be resource intensive. So it’s important to consider the following:
Storage. The workload that cloud delivers may be I/O intensive. So, organizations looking to deploy cloud computing must have a storage platform ready handle these types of heavier demands. Many IT departments look for solid-state drives or even embedded flash cards on a storage controller. By offloading resource intensive processes – boot-up, virus scans, or updates – administrators can control the number of dedicated IOPS they have to delivery per workload. Remember, without a good storage system, any workload running on that controller can be affected by the intensive demands of a large cloud platform.
Network. When deploying a distributed cloud model, there has to be a solid core networking infrastructure. In some cases, this means having multiple 10Gbe connections moving from a chassis into other racks. Without the right type of bandwidth, performance will lag and result in a very poor end-user experience. When designing the right type of architecture, look for unified systems where many of the necessary networking components are built in. Advanced fabric networking and good switching technologies are able to handle the type of workload demands that the modern user requires. Good planning around the network will not only benefit your cloud platform, but other IT components will see positives as well.
Hardware. In almost every case, a cloud model has to be able to scale and be agile. That’s why many organizations deploying such a solution have to think diligently as to the right type of platform to deploy. In some smaller deployments, rack-mount servers might work. However, for larger initiatives, a chassis platform with blade technologies is probably the right way to go. Capable of fast network connectivity, extreme scalability and lots of shared physical resources – chassis systems have become very popular for private cloud projects. In selecting the right hardware, ensure that you understand your virtual desktop and the resources that it will require. Also, make sure you note clearly the amount of users you plan on deploying per blade. Finally, and very importantly, always plan for cloud redundancy!
End-points. One of the biggest benefits of cloud computing is the ability to deliver the workload to a number of different types of end-points. In designing a cloud environment, it’s important to know the type of end-points that will be receiving the workload. In some cases thin-clients may be the right option. In other instances, cloud resources can be used to deliver users utilizing a BYOD policy. The point is that depending on the end-point, the location and the resources there have to be special considerations made. Ensure that policies are in place to create the best possible end-user experience. A poor cloud network will only frustrate the end-user and reduce productivity. In planning out the end-point, create the cloud use-case and you’ll quickly be able to see the right type of device necessary to make your platform experience optimal.
A big reason so many organizations jumped off the cloud bandwagon was because they forgot the all-important scoping and planning process. By first identifying a solid use-case and then understanding what the workload will be – your organization can take the right key steps in the project deployment process. When technologies and business visions align, cloud computing can have some excellent benefits for almost any organization.
By Bill Kleyman