We’re all familiar with the rise of cloud computing. Now, a new layer of networking is developing in the form of fog computing.
Fog computing – a middle ground between devices and the cloud – leverages intelligence at the edge for more efficient data handling and faster, local decision making. It collects and aggregates data from multiple sensors, and can do simple data analysis at the edge.
“We think that’s a real trend for the future,” says John Canosa, chief strategist at ThingWorx, a provider of M2M platforms.
ThingWorx sees M2M not as a bunch of sensors sending data into the cloud, he adds, but rather as a distributed computing solution in which even the sensors have some intelligence that can be leveraged to make decisions.
Cisco Systems (News - Alert) is generally credited with introducing the concept of fog computing, which it supports today via a technology called IOx that lives inside its Connected Grid Router products and will be expanded to other Cisco solutions in the future.
Todd Baker of Cisco explains that the value of fog computing in M2M scenarios is threefold. It acts as a data filter of sorts, making sure only the M2M sensor data that needs to be passed to the cloud is sent there. It serves as a simple control system at the network edge so if, for example, sensors detect an urgent need for adjustment on a remote oil pipeline, action can be taken locally for faster results. This can feature can also save valuable bandwidth. And it converts data sets, which may be based on specialized interfaces and protocols, at the edge so they are cloud ready once they ascend to that next network layer.
“It is getting very clear that the Internet of Things requires a different computing model, one that enables distributed processing of data with the level of resiliency, scale, speed, and mobility that is required to efficiently and effectively deliver the value that the data that is being generated can create when properly processed across the network,” Roberto De La Mora, Cisco’s senior director of Internet of Things products and solutions marketing, writes in the Cisco blog. “This distributed computing model is called fog.”
Like cloud, he writes, fog provides data, compute, storage, and application services to end users.
“The distinguishing characteristics of fog are its proximity to end users, its dense geographical distribution, and its support for mobility,” says De La Mora. “Services are hosted where they’re used: at the network edge or even end devices such as set-top boxes or access points. By hosting services locally, the fog paradigm reduces service latency and improves QoS, resulting in superior user experience.”
Paul Glynn, CEO of Davra Networks, which came out of stealth mode June 17, is also a proponent of fog computing for M2M. Davra provides the cloud-based RuBAN solution, an application enablement platform on which network VARS and system integrators can build services related to the Internet of Things. Cisco Senior Vice President Barry O’Sullivan joined the Davra board last summer.
“Our customers are using the cloud-based RuBAN platform to turn IoT raw data into usable everyday solutions that are as visual and easy to use as your smartphone,” Glynn says. “We work with our partners to do all the heavy lifting and installing of Cisco IoS Integrated Services routers and sensors on location, then we manage information coming from the IoT environment as well as information from the network itself. We make sure to bring only the relevant raw data back to the customer. That’s what makes fog computing so compelling, it’s about decision-making at ground level. Our VAR customers value our flexible solution because it can be used as easily in vehicle tracking or water meter usage as it can in smart cities and farms.”
Glynn provides a specific example of where fog computing might come into play. Say there’s a street with sensors for waste management, parking, and other purposes, on one street. Those sensors might connect over Wi-Fi to a router, and that router is their connection to the Internet. Fog computing sits in the router and makes local decisions about sensors, and sensors can talk to one another. A city Davra is working with in Europe has lots of underground rivers, which makes the streets icy when it gets cold, he says. So when the temperature drops and humidity is at a certain level, sensors know it, and the system sends an alert to the city council to deploy a truck to put sand or salt on the street.
Edited by Maurice Nagle