I am not sure if “Cloudification” is a legitimate word (I am sure it is not) nor I am inventing a new word! Please read-on to understand my view of introducing key cloud environments into the FogOS. Enjoy the reading.

The architecture of IoT networks involves a large number of end-devices aka “things” (e.g., Sensors, Actuators) connected to aggregating devices called Fog Nodes or Gateways, which in turn are connected to remote cloud IoT platforms. While the end-devices in IoT, may or may not have much processing resources, it is an architectural necessity for Fog Nodes to be built with decent capabilities. Thanks to their strategic location in the network and available processing power, Fog Nodes are well suited for providing a wide variety of conventional network services, message optimization, and local decision making which is extensively referred to as edge computing.

Traditionally FogNodes are used to provide basic network services. They are powered by Linux flavoured operating systems such as OpenWRT. Generationally, as the performance of the chipsets powering the Fog Node improves, many more interesting and necessary features are feasible to be implemented on the Fog Node that were only possible to deliver over cloud earlier. Some of these features take the form of an individual application by themselves, for example – Analytics Pre-processing & Data-Reduction logic, etc.  So, it makes sense to think of FogOS to be capable of hosting applications, sometimes natively and sometimes as on-demand basis.

Cloud Application Vs Embedded Application

Traditional application development on Gateways followed a very different philosophy compared to modern cloud-native application development. While previous method, relied on building software packages into Linux system and deployed them as an embedded software, cloud-environments evolved with containers and micro-services. This chasm needs to be bridged in the new FogNode OS. Our vision for Edge computing is to have cloud-native workloads to seamlessly run on FogNodes, with FogOS that is capable of having a high-throughput data-plane layer with support for multiple containers and virtual NICs.

Our flavour of FogOS at Vitalpointz is called VESPa-G (vitalpointz edge services platform agent) that is linux based, like most of the operating systems used in IoT. VESPa-G provides support for low memory, low compute, small footprint devices and systems with critical real-time response and requirements for precision in timing. VESPa-G is energy efficient and hardware agnostic. VESPa-G provides critical security controls to secure hardware and software and provides better Eco-system & ease of new application development. VESPa-G Supports a variety of wired (USB, Serial, Ethernet, etc.) and wireless (Wi-Fi, BT, ZigBee, LoRA, etc.) protocols, along with web protocols with low transmission overhead like CoAP, MQTT, or UDP alongside HTTP. In addition, VESPa-G provides end-device management, secure OTA software/firmware updates, provisioning, and two-way communication to the cloud and the devices.  We are embarking on a path where VESPa-G powered Fog Nodes when connected to vitalpointz’ IoT cloud platform, will enable workload migration from cloud infra to move seamlessly into FogNode and execute locally.