Fog calculation sometimes also called the Edge Computing is the process of performing the calculations locally and then passing the results to cloud processes.
The need arose when IoT devices entered the scene and when cloud systems started to be overwhelmed by processing RAW data on cloud computing resources. This has led to the need to process the raw data in local storage and with the computing power of IoT devices and send the processed data over the Internet to reduce cost and effort in terms of network usage, saving the cloud computing power and cloud storage.
All IoT devices generate terabytes of raw data from local sensors or transactions, sending everything to the cloud, the role of fog computing is to do as much processing as possible using compute units co-located in the devices data generators, so that
- Processed rather than raw data is transmitted.
- Bandwidth requirements are reduced.
- The latency between input and response is minimized.
- Preserve raw data where it will be used rather than dragging it back to the same device when needed.
Definition
“Fog Computing is a decentralized computing infrastructure in which data, compute, storage and applications are located somewhere between the data source and the cloud. Like edge computing, fog computing brings the benefits and power of the cloud closer to where data is created and operated. »
Fog Computing Architecture
Although not a separate system, it is a layer sandwiched between the cloud and the physical devices.
Implementing fog computing involves writing or porting IoT applications to the network edge for fog nodes using fog computing software, a fog computing program, or other tools. The nodes closest to the edge, or edge nodes, collect data from other edge devices such as routers or modems and then direct the data they receive to the optimal location for analysis.
By connecting fog and cloud networks, administrators will assess which data is most time-sensitive. The most critical time-sensitive data should be analyzed as close to where it is generated as possible, in verified control loops.
The system will then forward data that can wait longer to be analyzed to an aggregation node. The characteristics of the fog calculation simply dictate that each data type determines which fog node is the ideal location for analysis, depending on the ultimate goals of the analysis, the type of data, and the immediate needs of the user. .
Advantages and disadvantages of Fog Computing
Advantages
- Less network traffic: Fog computing reduces traffic between IoT devices and the cloud.
- Offline availability: In a fog computing architecture, IoT devices are also available offline.
- Cost savings through the use of third-party networks: Network providers incur high costs for high-speed uploading to the cloud. Fog Computing reduces them.
- Data Security: When fogging, device data is pre-processed by the local network. This allows for an implementation where sensitive data can remain internal to the company or be encrypted or anonymized before being uploaded to the cloud.
Disadvantages
- Additional Network Security Requirements: Fog Computing is vulnerable to man-in-the-middle attacks
- Increased maintenance requirements: Decentralized data processing requires more maintenance because controllers and storage locations are distributed across the network and, unlike cloud solutions, cannot be maintained or administered centrally.
- Little protection against failure or misuse: Companies that rely on fog computing must equip IoT devices and sensors with controllers that are difficult to secure against failure or misuse, for example in manufacturing facilities at the network edge.
- Higher hardware costs: Fog computing requires IoT devices and sensors to be equipped with additional processing units to enable local data processing and device-to-device communication.
Fog Computing in banking
Fog Computing is the distributed processing medium in payments, recommending personalized recommendations and offers and attracting a new generation of customers through groundbreaking payment methods like Apple Pay, Samsung Pay any other unrestricted on-device financial transaction payments but can be extended to risk assessment. and trading platforms.
There are several use cases where fog computing has become an integral part of implementing functionality for various financial institutes around the world. Some examples but not limited to:
- Citibank uses beacon technology to allow consumers to access ATMs using their smartphones.
- Small financial institutes are building the analysis and analytics algorithms to run on local devices rather than creating expensive cloud-based solutions
- Provide insurers with insight into real-time driver driving habits and vehicle status.
Today’s highly competitive banking industry, driven in part by the rapid growth of new IT paradigms, as well as financial technology (Fintech), is driving the industry to seek ways to continue to improve customer relationships. Analytical processes in cloud environments can leverage large volumes of data to perform computational processing, including machine learning techniques to improve reliability, automated configuration, and performance
In e-business, one way to do this is to offer personalized product recommendations. Banks are participating in content personalization methods to grow and align with new digital business mechanisms. In digital businesses, recommender systems provide users with intelligent mechanisms for finding products tailored to their preferences. The increase in sales of this type of systems is a consequence of their ability to interact with users to help them choose and discover the products and services that interest them. In this sense, recommender systems are designed to adapt to each user, becoming a sort of personalized assistant that facilitates access to the many product offers in a more efficient way.
Fog computing can be used to present predictions in areas of financial products, such as mortgages, loans, pension plans, etc.
Diagram of the Fog Computing architecture