Build the right hyper-converged cloud for your apps
Picking the right hyper-converged hardware and software for your application workload requires understanding how the elements of HCI can best be combined for your needs.
As companies seek to make their data centers operate more like the cloud, hyper-converged infrastructure has become increasingly popular. HCI's mixture of hardware and software forms the foundation for a hyper-converged cloud platform.
HCI and the vendors that supply it evolve as customers' application needs change. There are some things you must know to implement an HCI-based private cloud that best suits your application workload. The perfect private hyper-converged cloud platform requires a number of elements.
The hypervisor's role
Virtualization is a key element in any hyper-converged cloud. Some HCI appliance vendors only support VMware vSphere, whereas other vendors include a hypervisor in the form of the open source Kernel-based Virtual Machine (KVM). For example, Nutanix makes available its Acropolis Hypervisor, although it also supports VMware's vSphere hypervisor. Other companies, such as Scale Computing and Cloudistics, use KVM hypervisors, which customers can't modify, as the core of their offerings.
VMware's vSphere is the dominant software, so consider your other options carefully. Choosing a vendor that requires you to exit the VMware environment doesn't necessarily mean giving up features. KVM is incredibly powerful. But it does mean that you won't be able to participate in the large VMware ecosystem; some products will no longer be available.
An automation and orchestration engine
It's that mixture of hardware and software that forms the foundation for a hyper-converged cloud platform.
Automation drives the cloud, and a hyper-converged cloud is no different. Creating new workloads at Amazon doesn't mean an Amazon employee actually carries out the request. The power lies in the automation Amazon provides. Likewise, for a private hyper-converged cloud environment, it's important to be able to automate as many routine processes as is reasonable, with the goal being to drive efficiency.
Automation helps with singular tasks, whereas orchestration brings a new level of capability. Orchestration means stitching together multiple tasks into a single outcome. It also brings a level of management to the automation, which allows you to control the various scripts that enable automation. Using an orchestration engine helps keep any cloud platform -- public or private hyper-converged cloud -- under control.
User self-service capabilities
Using the automation that a cloud environment provides means your organization won't have to hire IT staff to perform routine operations. To adopt a private hyper-converged cloud, you need a platform that can provide some semblance of self-service so users can have routine requests for infrastructure and applications handled automatically.
Integration opportunities
Your organization probably already uses some type of cloud service, whether that's something from Amazon or something like Office 365. With that in mind, the private hyper-converged cloud platform you select must be able to easily integrate with the off-site cloud services the customer also uses.
