The term ‘cloud’ is one that is applied to a collection of web-based resources that can be used either for processing data with the use of pooled computing power, or for the storage of data. In the enterprise market, there is a new term coming forward – the ‘hyperconverged cloud’. Hyperconvergence is described as being a software-based infrastructure that has full governance of the multiple elements that provide the underlying cloud platform.
A cloud environment is constructed out of several individual elements that come together to provide a single unified environment on which cloud devices can be hosted. It is these elements that the platform software will have full governance of in a hyperconverged environment. The four main cloud components consist of:
- Compute – in the cloud, the compute devices are the servers that cloud virtual machines are running on; a compute server provides the CPU power and the RAM that virtual machines require to run services and process data – as a cloud often features a number of compute servers in the name of redundancy, where one compute server fails, the virtual machines hosted on that server can be relocated to an available compute server
- Storage – cloud data needs to be hosted somewhere, and in a standard enterprise cloud design, this is something that will normally be taken care of by a SAN (Storage Area Network) device, which is a network-connected device that contains a large number of hard drives that are arranged as to provide a high level of performance whilst safeguarding data; SAN devices can be accessed by compute devices through the storage network, therefore cloud virtual machines can be transported across compute nodes, as dictated by resource availability, without incurring any downtime or data loss
- Networking – the compute devices and storage devices need a way of communicating with one another, which is where the network element comes into play; to provide a robust and secure environment, external traffic is routed through a network that is external to the cloud – the sensitive data that is hosted by cloud virtual machines is instead routed through dedicated internal networks, such as a Storage Area Network (SAN) that will run between the compute servers and the storage devices
- Virtualisation – the virtualisation element is the software that runs on top of the underlying hardware; the cloud software is responsible for managing the allocation of resources and is the platform through which the compute, storage and networking elements communicate with one another.
Cloud hyperconvergence can be described as being a cloud infrastructure where the software powering the platform tightly integrates all of these elements and technologies from other sources together to provide a single solution that is supported through the service provider. Hyperconvergence allows for the rapid development of cloud environments by enabling service vendors to provision additional resources in the cloud with the use of physical hardware in no time at all, without there being any knock on effect on the performance of the existing cloud setup.