DED9

What is Edge Computing?

This article has tried to fully introduce edge Computing and point out some of its important applications.

Let’s introduce you to “edge” Computing. Edge is an up-to-date word. Like “IoT” and before that “cloud,” the edge means everything and nothing. But we’ve watched the presentations of some industry experts, listened to podcasts, and sometimes even read articles on the subject. And we think a useful definition and introduction of some possible applications for this up-to-date technology is needed.

What is Edge Computing?

In the beginning, there was a big computer. Then, in the age of Unix, we learned how to connect to a computer using dumb terminals (not in a derogatory sense). Next, we had PCs, and this was the first time ordinary people had the hardware that worked.

We are now in the age of cloud computing. Many of us still have PCs, but most use them to access centralized services such as Dropbox, Gmail, Office 365, and Slack. In addition, devices like Amazon Echo, Google Chromecast, and Apple TV support cloud-based content and intelligence – unlike the Little House DVD box in Prairie or the Encarta CD-ROM that you might enjoy in the age of personal computing…

As focused as all of this may seem, the amazing thing about cloud computing is that a large percentage of all companies in the world now rely on infrastructure, hosting, machine learning, and computing power for a small number of cloud providers: Amazon, Microsoft, Google, and IBM.

Amazon, one of the largest providers of this public cloud (compared to “private clouds” hosted by Apple, Facebook, and Dropbox), had 47% of the market in the last three years.

The advent of edge computing as a buzzword that you may have noticed is that it is being realized by these companies that there is not much growth left in the cloud. Almost anything that can be focused is focused. Most new opportunities for the “cloud” are now on the “edge.”

So, what is the edge?

The word edge in this context means real geographical distribution. Edge computation is performed at or near a data source, rather than relying on the cloud in one of the dozens of data centers to do everything. This does not mean that the cloud disappears. This means that the cloud is coming towards you.

Accordingly, let’s try to examine what people mean by praising edge Computing.

Delay

One of the most important stimuli for calculating the edge is the speed of light. If computer A has to request computer B on the other hemisphere before it can do anything, computer user A will perceive the delay as a delay. The short delay after clicking the link before your web browser starts showing anything is largely due to the speed of light.

Multiplayer video games implement some sophisticated techniques to reduce the actual. And perceived latency between shooting at someone and knowing with certainty that you missed a shot.

Voice assistants usually have to analyze your requests in the cloud, and the round-trip time can be very significant.

Your echo must process your speech send a compressed view of it to the cloud, and the cloud must decompress that image and process it – which may require ping. Another API is somewhere (to call another computer on the network that is on the network and test the action to see if we have sent or received or not), perhaps to understand the weather and add more speed Optical Delay.

And then the cloud sends you an echo response, and you can finally figure out that you should expect a maximum temperature of 30 and a minimum of 6 degrees Celsius today. And adjust your work schedule and coverage accordingly. Suppose it is to be done as an edge calculation. In that case, the whole operation will be done with less delay and more savings in network resources because many operations will be performed on the nearest computer on the network instead of in the cloud.

Definition

One definition of edge computation is any computer program that gets closer to requests with less latency. At an IEEE DAC 2014 Keynote and subsequently in a guest lecture at the MTL MIT seminar, Karim Arabi considered edge computing as widely as all out-of-cloud computing at the network edge, particularly in applications that process data as Immediately required to be defined. In his definition, cloud computing operates on big data, while edge computing operates on “instant data,” which is real-time data generated by sensors or users.

This term is often used synonymously with fog computing.

According to the State of the Edge, edge computing focuses on servers “adjacent to the last mile network.” “Anything that is not a conventional data center can be an edge for anyone.”

The edge knots used to play the game are called game lets, usually one or two jumps away from the client. “The edge node is typically one or two jumps away from the mobile client to overcome response time constraints for real-time gaming compared to cloud gaming,” says Per Anand and Edwin.

Edge computing may use virtualization technology to facilitate the deployment. And the execution of a wide range of applications on edge servers.

Concept

World data will increase by 61% to 175 zettabytes by 2025. The proliferation of IoT devices on the edge of the network produces huge amounts of data – storing and using all of this data in cloud data centers maximizes network bandwidth requirements. Despite advances in network technology, data centers cannot guarantee acceptable transmission rates and response times, which are often necessary for many applications. In addition, peripherals constantly use cloud data, forcing companies to decentralize data storage and service delivery and use physical proximity to the end-user.

Similarly, edge computing aims to move computing from data centers to the network’s edge. And to use smart objects, cell phones, or network gateways to perform tasks and provide services on behalf of the cloud.

Moving services to the edge makes it possible to provide content storage, service delivery, and persistent data storage. And IoT management, which increases response time and transfer rate. At the same time, the distribution of logic in different nodes of the network introduces new issues and challenges.

Privacy and security

The distributed nature of this template represents a change in the security plans used in cloud computing. In edge computing, data may travel between different distributed nodes connected over the Internet. And this, therefore, requires special cloud-independent encryption mechanisms.

Edge nodes may also be resource-limited devices that limit choice in terms of security practices. In addition, the shift from a top-down, centralized infrastructure to a decentralized trust model is needed. On the other hand, by keeping and processing data at the edge, privacy security can be increased by minimizing the transfer of cloud-sensitive information. In addition, ownership of the data collected shifts from service providers to end-users.

Scalability

Scalability in a distributed network has to deal with a variety of issues. First, one must consider the heterogeneity of the devices, the different functions, energy constraints, the highly dynamic conditions. And the reliability of the connections compared to the stronger cloud data center infrastructure. In addition, security requirements may further delay communication between nodes, slowing down the scaling process.

Reliability

Breakdown management is critical to keeping services alive. If a node shrinks and is not accessible, users should still access a service without interruption. In addition, edge calculation systems must take steps to recover defects and alert the user to an accident.

For this purpose, each device must maintain the network topology of the entire distributed system. So that error detection and recovery can be easily performed. Other factors that may affect this aspect include the connection technologies used, offering varying degrees of reliability. And the accuracy of the data generated at the edge is unreliable due to specific environmental conditions. For example, an edge computing device, such as a voice assistant, may provide local users even during cloud services or Internet outages.

Speed

Edge computing brings analytical computing resources closer to end-users and can, therefore, increase applications’ responsiveness and power. A well-designed edge platform performs significantly better than a traditional cloud-based system. Some applications rely on short response times, making edge computing significantly more efficient than cloud computing.

Examples range from the Internet of Things to self-driving, anything related to human/public health or safety, or involving human perception such as face recognition. Which typically takes between 370 and 620 milliseconds for humans to do. Edge computing can most likely mimic the same speed of human perception. This is useful in applications such as augmented reality, where the headset should preferably recognize who the person is simultaneous as the user.

Efficiency

Due to the proximity of analytical resources to end-users, sophisticated analytical tools. And artificial intelligence tools can be implemented at the edge of the system. This edge placement helps to increase performance and has many system benefits.

In addition, I am using edge computing as an intermediate step between client devices and the wider Internet results in cost-effective savings. As shown in the following example: Using servers located on a local network to perform computations, video files only need to be transferred over the local network. Avoiding transmission over the Internet significantly saves bandwidth and thus increases performance. Another example is voice recognition. Only known text can be sent to the cloud if detected locally instead of recording audio. Thus significantly reducing the amount of bandwidth required.

Applications

Edge application services reduce the amount of data that must be moved, the resulting traffic, and the data’s distance. This will reduce delays and reduce transfer costs. Computational loading for real-time applications, such as face recognition algorithms, showed significant improvements in response time, as shown in the initial research.

Further research has shown that using resource-rich machines called cloudlets near mobile users. Which provide routine cloud services improve some tasks’ execution time at the edge node. On the other hand, the loading of each task may be slowed down due to the transfer time between the device and the nodes. So an optimal configuration can be defined depending on the workload.

Edge computations have created countless examples and applications in the real world, including:

This technology has numerous other uses, including improving health care, transportation, and retail jobs. Other notable applications include connected cars, self-driving cars, smart cities, Industry 4.0 (smart industry), and home automation systems.

Die mobile Version verlassen