Quanta Cloud Technology (QCT), a provider of servers, has allied with Robin.io to create an integrated platform for running containerized applications on Kubernetes clusters deployed at the wireless 5G edge.
Mehran Hadipour, vice president of business development and technology alliances for Robin.io, says the IronCloud — Robin Cloud Platform is based on the Multi-Data Center Automation Platform (MDCAP) and Cloud Native Platform (CNP) that Robin.io created. Robin.io packaged its own distribution of Kubernetes with its software-defined storage platform to run stateful containerized applications.
The goal is to make it easier for service providers and large enterprises to employ carrier-grade bare metal 1u and 2u platforms to deliver additional services to the network edge. While most carriers today are rapidly rolling out 5G connectivity services, they need to be able to deliver additional application services to make investing in 5G worthwhile, notes Hadipour.
Based on third-generation Intel Xeon processors, the servers provided by QCT will make it possible for organizations to deploy a hardware platform at the edge of the network designed specifically for those environments.
In addition to being able to run Kubernetes on a bare metal platform, Robin.io also makes it possible to run legacy applications designed for virtual machines on top of Kubernetes. The Robin.io distribution is unique in that it enables virtual machines to run on Kubernetes using a layer of software that eliminates the need for a hypervisor, says Hadipour. In effect, Kubernetes becomes the hypervisor for the virtual machine using a layer of software developed by Robin.io, adds Hadipour.
Bringing application code to the network edge is becoming more critical because these applications typically require data to be processed and analyzed as closely as possible to the point where that data is created and consumed. The most efficient way to build, deploy and update applications running on the edge is to encapsulate the microservices employed to build them within containers.
There are, of course, going to be a wide variety of edge computing applications spanning everything from mobile computing to Internet of Things (IoT) deployments on factory floors. Each of those applications is going to require access to servers in relatively close proximity to drive event-driven processes in near-real-time.
It may be some time before service providers have the necessary level of infrastructure in place to deliver more than just additional bandwidth. However, there may soon come a day when there is more application code running at the edge than in the cloud. That doesn’t mean cloud computing platforms will become obsolete; rather, the relationship between the edge and the cloud will evolve to the point where not as much raw data will need to be processed in the cloud.
The cloud will, in effect, become the repository for aggregated data that has first been processed and analyzed at the edge. The goal, of course, is to reduce the amount of raw data being transferred across wide area networks.