Edge computing and fog computing are relatively new concepts that expand on the cloud metaphor. While cloud activity takes place at a distant location (“away in the clouds”), such as a processing center in a different country, edge activity happens closer to the user’s location (at the “edge of the cloud”), such as in a home or office environment, commonly involving Internet of Things (IoT) devices. Fog activity happens somewhere in between.
Benefits of edge computing include the ability to perform tasks more quickly, as applications don’t have to wait for information to be transferred to and from a distant cloud location. For example, an automated security system, perhaps equipped with custom software, might locally analyze video for instances of movement, sending only those that include it to an offsite (cloud) location for further processing.
However, edge devices and systems are often manufactured without as much thought given to security as within remote cloud processing locations. This situation creates the potential for hackers to more easily gain access to edge devices as well as the more important company networks they may be connected to. Because of their more robust security features, fog computing systems bring the possibility of greater security to edge applications.
Benefits of Fog Computing
Fog computing is an extension of cloud computing. It offers the power of cloud resources but brings those resources closer to the IoT devices that use them. In this way, fog computing enables the immediate computing needed by edge devices employed in time-sensitive scenarios.
For example, a surgeon using a robotic device to help perform surgery may require feedback in less time than it would take for that data to travel from a cloud network. The same is true for driverless cars or smart traffic lights that may require immediate information to make a split-second decision that could save a life.
Other examples include applications that aren’t necessarily life-or-death but could benefit from faster data transfer and lower latency, such as smart utilities, which use real-time data to provide more efficient services for citizens. In any of these scenarios, information can be processed much more quickly by a fog network than by a cloud one.
None of this is to say that fog networks should replace those in the cloud. Each can be a critical component of an efficient overall computing strategy. The fog is perfect for short-term computing tasks, while the cloud is better for intensive work requiring massive data sets or storage. Companies and organizations are smart to use both for different applications.
Why Security Is an Issue
Cloud server farms are typically housed within highly secure data centers that include robust firewalls and network monitoring and architecture. Conversely, both edge and fog devices and systems rely on “infrastructure [that] needs a flexible, distributed security model, with easy-to-add-and-remove edge devices, and easy-to-relocate fog servers,” according to an SC Media article. This distributed configuration makes a coordinated attack possible, potentially taking out several vulnerable nodes at one time.
The article notes that the biggest threats to these less secure resources are “denial of service attacks, man-in-the-middle attacks, and rogue gateways. There are also the added risks of misuse of resources, privacy leakage, virtual machine manipulation, and injection of information.”
How the Fog Can Help
The fog provides a middle ground between the edge and the cloud. It holds the potential to include some of the security measures that edge systems lack and reduced time for data to travel. Data can be transferred from edge devices to more secure fog computing systems, providing the security offered by cloud resources, but with greater speed. To ensure better protection for more sensitive data, fog systems can be segmented to define different privileges for different parts of the network.
While data sent to the fog can provide greater security than edge devices, it can also provide superior security over the cloud. When sensitive data is moved, there is a higher chance it could be compromised. Keeping data that doesn’t need further processing housed more locally, such as within a fog network, keeps it under the owner’s control.
Different tasks require different types of computing. Local networks and storage are appropriate for processing small amounts of non-sensitive data, while cloud resources are useful for storing large amounts of data or performing complex processing tasks. These processing types have fulfilled the computing needs of individuals and businesses for many years.
However, the emergence of IoT has created new requirements. IoT devices, including sensors and manufacturing tools, must be able to operate on their own to provide the intended benefits of transferring detailed data to larger computing systems (such as those in the cloud). While highly convenient, this edge interface is also inherently insecure. Using a fog configuration, IoT users can maintain the convenience factor while ensuring their edge devices remain more secure.