Since the early 2000s, cloud computing has played a dominant role in enabling the many applications we rely on in our daily lives and business operations. However, as the internet of things continues to gain traction, cloud computing alone becomes increasingly inadequate. Many emerging applications require services a bit closer the edge, and can’t wait for data to travel all the way to the cloud and back. The discussion has evolved from simply determining how fog computing will fill the technology gaps. We know fog computing has to play a role in an IoT world, and so we need to start talking about the profound business opportunities and disruptions fog computing will bring.
Fog computing is already influencing how edge networks are being built. Routers, switches, wireless access points, application servers and storage servers at the edge are already converging into unified “fog nodes.” These often feature a unified networking platform that supports heterogeneous networking technologies, and a common computing platform that supports applications from multiple suppliers. Applications from different providers will no longer need their siloed hardware and software platforms. This ongoing convergence of computing, networking and storage at the edge will significantly reduce system complexity and cost, increases system and application manageability, and make it easier for applications to interact with each other. The convergence of networking and computing at the edge will also pave ways for the integration of operational technologies and information technologies.
It isn’t an either/or with fog and the cloud. On the contrary, fog empowers the cloud by connecting a vast range of devices to the cloud. Connecting all potential IoT devices directly to the cloud will often prove impractical due to limited abilities on many IoT devices, excessive complexity and cost to add cloud connectivity management to all IoT devices, and scalability limitations imposed by connecting every device directly to the cloud. Fog can use simpler local procedures and protocols to interact with the devices and shield the devices from the complexity of direct interactions with the cloud. Furthermore, fog can act as the proxy of the cloud to deliver cloud services to IoT devices and systems. For example, a fog node on a connected car can act as the proxy for the tens of microcomputers on the vehicle. Instead of requiring every onboard microcomputer to directly connect to the cloud for software updates, the fog node can retrieve the software update packages from the cloud and then install the updates on the microcomputers at the right times (e.g., when the vehicle is parked and not expected to be used for long enough to complete the software update process).
Fog enables many new services that the cloud alone cannot effectively support. For example, a fog system, more powerful than the endpoints, can provide local security services. Such security services may include local monitoring of the endpoints’ security status, security credential and software updates to the endpoints, and malware detection and protection on behalf of the endpoints.
Fog as a service will allow users to access private and public fog systems deployed close to them. Customers can store their data and host their applications in these nearby fog systems. A manufacturer, for example, can store its sensitive data in a private fog system deployed close to its factories. It can also have its manufacturing control applications hosted by local fog systems. Customers may simply rent storage spaces or computing servers from these fog systems and manage their data and applications by themselves. Customers could even receive turnkey services from a fog-as-a-service operator.
Fog will allow players of all sizes to deploy and operate fog systems and services. Many new, local and regional fog operators will emerge, in a way similar to the rise of local and regional Wi-Fi service operators. This will further disrupt the existing centralized cloud business models.
So far, we’ve seen fragmented efforts that tend to overly emphasize the uniqueness of computing at the edge and overlook the importance of seamlessly integrating computing at the edge with the overall end-to-end computing ecosystems. The results are siloed systems for different networks and industry verticals, and even siloed applications inside each industry vertical. Customers could be forced to use different edge computing platforms for 5G, wired telecom networks, enterprise networks and even for different application domains such as manufacturing, smart cities and smart grids.
For fog to be successful, and ultimately for IoT to be its most impactful, the industry needs a common platform that can work for multiple industry verticals. Organizations like the OpenFog Consortium are currently developing such horizontal architectures. This vision is analogous to that of TCP/IP. While TCP/IP provides a horizontal platform for distributing data packets from place to place, the horizontal fog platform will provide a horizontal platform for distributing computing functions from one place to another, plus additional capabilities for pooling the distributed resources to support user needs and for managing and securing the distributed resources.
This horizontal fog platform will work not just over but also inside any wireless and wired network. For example, the same fog platform will work inside a 5G wireless system to allow telecommunication service providers to move advanced network control functions and user applications into the radio access networks, as supporting such mobile edge computing requires the same core capabilities of a general-purpose fog platform. These capabilities include distributing computing functions from one place to another, pooling and orchestrating computing resources in different places to support end-to-end applications, managing the lifecycles of the computing functions and resources, supporting applications from different providers (Multi-tenancy), and securing the distributed resources and applications. This will also solve the silo problem, and customers can invest in one platform instead of multiple siloed ones.
IoT presents many unique challenges, but none is more important than solving the infrastructure challenges. Fog computing will play a vital role, but as it begins to fill current technology gaps, it also presents new and exciting opportunities for technology players who can help customers fully embrace IoT.
All IoT Agenda network contributors are responsible for the content and accuracy of their posts. Opinions are of the writers and do not necessarily convey the thoughts of IoT Agenda.