1. What is Edge Computing?
Edge computing is a software architecture in which includes processing of client data such that processing takes place as close as possible to the initiating source. In simpler terms, rather than processing and analyzing the raw data at center, edge computing shifts some part of storage and process them out of fundamental data center and closer to the source from which it has originated. Thus, edge computing brings data processing in close proximity to the data source. This helps in reshaping IT industries and business domains.
2.Working of Edge Computing
In a system with traditional network design, the required data is produced at user’s end. The data gets communicated across Wide Area Network (WAN) by connecting Local Area Network (LAN). After processing the data, the results are sent back to the user or client source.
But this method is only effective when data processing and storage takes place in a small area specified by WAN. With the growing businesses all over the world, the number of devices connected to the internet have increased. Since these devices are not limited to a specific area and have spread worldwide, we cannot use WAN network communication here.
This is where edge computing technology comes into picture. The computing equipment of edge computing is usually protected from external environmental conditions such temperature and moisture. Once the data is sent for processing, it is analyzed with business point of view. The processed results are sent back to the source from which the data was initiated.
IT industries use edge infrastructure to store and process data. Since the data has to travel across the globe, if we can’t bring required data to data center, we build a data center near to the data boundary where it can get processed and stored. It is same as open new branches of a single company in multiple locations rather than depending on a specific location. This increases efficiency of work resulting better outcomes.
3.Comparison between Edge, Cloud and Fog Computing
Fog computing and edge computing are sometimes used interchangeably and share an almost similar architecture. But it has to be noticed that these terms are different to each other.
Edge Computing:
Edge computing is the positioning of operating and storage resources at client’s location. Hence computation and storage of data takes place at the same place from where data is originated. This can be understood with an example of railway station data and operation management. The rail traffic sensor data is computed and directed across various railway stations for wide-ranging analysis.
Cloud Computing:
Cloud computing is highly scalable distribution of computation and data storage without direct supervision by the user and is instead stored in an amorphous cloud. Cloud computing uses ‘pay-as-you-go’ model which helps the company to reduce its storage expenses as it has to pay only for the resources which are in use and not the entire package with which the cloud comes.
The main aim of cloud computing is that the users should be able to work on all the technologies associated with it and reducing the cost by focusing on their core business. Could computing is majorly used for Internet of Things (IOT) in which many devices are connected across the network and hence requires major amount of space for data storage. Unlike edge computing, cloud computing does not allow data to exist closer to the data sources.
Fog Computing:
Fog computing is a distributed computing architecture in which data is processed and stored between data source and cloud. It uses IoT sensors, cameras and other devices. Fog computing and edge computing both are used for storage but fog computing is closer to client and is spread over wider geographical location than edge computing.
It is mostly used in smart cities and smart buildings. Smart cities use fog computing for public transportation system and urban planning. Fog nodes are present at certain distances for better collection and processing of data.
4.Why is edge computing required?
Since a lot of devices are involved in computing tasks, the edge computing architecture supports distributed computing. But decentralization of data requires higher level of control and monitoring. Edge computing solves three major challenges associated to network architecture. These are discussed below:
- Bandwidth
It is the maximum amount of data that can be transferred across a network connection. It determines quality and speed of the connection. A limited bandwidth indicates that only a minimum number of devices can be connected across the network. If the bandwidth is increased, the cost of network system increases.
- Overcrowding
Since all the devices are connected using the internet, and processing such are file streaming and data exchange takes place, overcrowding takes place. This forces data to consume more time during transmission and retransmission.
- Expectancy
Even though data is transmitted at the speed of light, when it has to travel to long distances, the speed decreases by some amount. This results in delayed response time when functioning with real time data. By using edge computing, as the data is operated in smaller intervals of LANs, data gets transferred efficiently even with less bandwidth, hence reducing the cost. This also eliminates the problems related to overcrowding and expectancy. Since the data is stored locally, data is pre-processed and decisions are made in real time without delay.
6.Applications of Edge Computing
- Business
Business includes retail shops which produce stocks and sell them. It also involves other real-time businesses. Edge computing helps in better analysis of the diverse stock and predict sales. This turns out to be an effective solution to increase sales across small retailers.
- Transportation
Smart vehicles are required to collect the information about the route and location to which they are travelling, their speed, traffic conditions and condition of other vehicles. Since they work on real time basis, a significant and fast computation takes place. Edge computing helps in this as each travelling vehicle acts as an edge with specific data related to it.
- Agribusiness
Sensors and various IoT devices are used in a controlled farming. Edge computing is used to keep a track of favourable conditions of each type of crop. Other conditions include water usage, nutrients required and harvesting conditions. Data is collected for better analysis of crop to improve its harvest.
- Healthcare Industry
Patient’s data is collected from smart equipment used in healthcare industries which work on sensor data. Since the data is huge and complex, we use edge computing to apply machine learning algorithms to predict real time results from it. The data is stored in a cloud and can be accessed whenever required.
Challenges encountered with edge computing
- Authentication
In edge computing, data travels through various devices and distributed nodes. Since all the IoT devices are connected to the cloud, safety of data is one of the most important things. To avoid hacking or misuse or data by the hackers, regular updates, software patches and bug fixing have to be taken care of. To secure the data, it has to be properly encrypted so that it does fall into wrong hands which might use the personal data inappropriately.
- Connectivity
Even though everything on edge computing is connected to cloud, it always requires a minimum hardware system to account for some major connections. This creates a problem as it is difficult to design a single system which handles all the important connections that are required for a proper edge computing system.
- Manual Maintenance
Even tough all the devices are connected through internet, whenever any fault occurs, they need to be checked physically. Maintenance and replacement of equipment is always present during system failure. This might occur due to devise used in Internet of Things (IoT) with limited lifespans.
- Lifecycles
Lifecycle of a data is another important factor in edge computing technology. Every produced data has a certain time to which it can be used and becomes redundant after that. For example, a business has a lifecycle to its data which cannot be used after certain period of time. Sustaining such unwanted data increases the storage capacity redundantly and the real required data gets congested while storing.
Read on Extended Reality A Future Habitat Version Machine. Click here
