A hyper-converged infrastructure tightly integrates storage, compute, network and server virtualization resources, sometimes from different companies, in a single box, which has the ability to serve as the basis of a scale-out system. What is so exciting about an integrated stack that it has become the next big thing in data storage?
Like the hyper-converged infrastructure itself, it is a convergence of many aspects working in concert: scalability, resilience, simplification, ease of deployment, and streamlining.
Hyper-converged systems are modular and designed to scale out by adding additional nodes to the base unit. The more appliances added, the greater the performance and capacity of the structure. This is more efficient than scaling up by adding more drives, memory, or CPUs on an individual basis.
The only drawback to this is that in a vendor-supplied system it is not possible to make small upgrades or tweak it -- an entire additional appliance is required. In a build-your-own platform, however, it is possible to select your preferred drive configuration and server hardware manufacturer, as long as it is supported by the vendor you are using.
Hyper-converged infrastructures are also scalable in that they can be used by any sized business organization to meet its overall computational needs.
Because of the appliance design, hyper-converged infrastructures have no downtime during planned maintenance, or in the event of disk, network, or host failures.
Instead of having an IT team and a set of applications to manage each discrete aspect of your array, a hyper-converged infrastructure can be maintained by a single team or person. This has the additional benefit of freeing up the IT department to bend its energies to greater innovation for the business organization as a whole. If your IT department is already on the lean side, then hyper-converged infrastructures can be a boon, as the vendor handles the maintenance and the smaller IT department (or person) does not have to be all things to all servers.
Hyper-converged infrastructures are supported by a single vendor, which greatly simplifies the upgrade process and can reduce maintenance costs.
Ease of deployment
The hyper-converged infrastructure integrates all the functions of a typical converged server stack, which radically simplifies the deployment process by making it as close to plug-and-play as it is possible to get.
Most IT departments use a converged infrastructure, which combine two or more separate components brought together to function cooperatively with the goal of minimizing compatibility issues and simplifying management. Hyper-converged infrastructures take this one step further by seamlessly combining hardware, software, and support in one tightly integrated package. While this is not meant to replace the typical converged infrastructure, it can serve to complement it by expanding its functionality through scalability. Hyper-converged infrastructure appliances allow the integrated technologies to be managed as a single system through a common toolset.
Hyper-converged systems bring the advantages of simple design, decreased administrative overhead, and simplified vendor management to highly virtualized environments. They can also expand beyond compute, storage, networking, and virtualization to include other technologies such as WAN optimization, data deduplication, snapshot capability, and inline compression.