Ask yourself this question before reading this article
: “Why do I need to use edge computing?”



As more and more tech companies claim they offer
The ability to “edge intelligence”, or edge computing, has attracted the attention of many executives. What company wouldn’t want more intelligence? However, deciding whether an edge computing strategy is truly beneficial means understanding which applications benefit from edge computing and how to build networks to support edge services.

Let’s start with this question
: “Why do I need to use edge computing?”

Applications that suffer from latency or other network-related problems may benefit from edge computing. This is because edge computing puts high-performance computing, storage, and networking resources as close to end users and devices as possible. While improving application performance for end users is a key attribute of edge computing, there are other advantages to edge computing
:

Follow the rules and regulations
— Keep the data within a certain geographic range.

security
Perform authentication and other security features closer to the end user.

The elastic
Leverage distributed cloud resources to reduce dependence on applications that reside in a single (or two) cloud region.

Don’t misunderstand the intent of brinkmanship
— It’s not about replacing cloud providers like Amazon or Microsoft; It’s about building those capabilities. When and how to enhance the cloud will be based on an in-depth understanding of the best places for users and applications to interact.

Which edge are we talking about?

Delving into edge computing can be a little confusing at first. On multiple access edge computing

(MEC), a set of standards built on mobile networks. Fog computing is a frequently used term in the industrial Internet of Things, but edge computing strategies do not necessarily have to be tied to the implementation of the Internet of Things.

In both cases, one might conclude
: You think you don’t need to use marginal strategy, but you might actually need to use one or two different edges. The Linux Foundation provides a glossary for guidance. The edge terms that enterprises should be familiar with are: infrastructure edge, access edge, and aggregation edge.

Infrastructure edge
: “Edge computing… Deployed on the carrier end of the last mile network. Computing, data storage, and network resources at the edge of the infrastructure support cloud-like capabilities.

Both the access edge and the aggregation edge are sub-layers of the infrastructure edge. The access edge is closest to the end user or device, and the aggregation edge sublayer is another hop farther from the access edge. This edge may consist of medium-sized data centers in a single location or of multiple interconnected microdata centers.

Edge of the network

Network is an often overlooked element of edge computing architecture. Whether your brinkmanship is to move existing applications closer to the end user or to create new ones, you need an agile network to bring all the elements together.

The first step toward a ready-to-use network is to use a flat network architecture to move traffic from a regional branch of an enterprise or partner and supplier to a regional center and aggregate traffic to a regional center. These hubs are located in carrier-neutral multi-tenant data centers that provide points of interconnect with cloud service providers. Hubs access the network through various means
(MPLS, broadband, wireless) connected together, but managed as a logical network via SD-WAN technology. Creating overlay networks through software makes it easier to configure and tune routing and application level optimization.

in
In the Linux Foundation’s terminology, this interconnect hub architecture can also be called converged edge.

How do I use edges now

After distinguishing these different layers and the means of connecting them, which one serves the goal of enhancing existing enterprise applications
? In many cases, the convergence edge will be the target of choice for the enterprise. Some ways to take advantage of aggregate edges include:
Move the security element closer together
“Attack

The deployment of
“Source of enterprise firewalls and move access control closer to end users to improve user experience and reduce bandwidth requirements for core WAN links.

User requests are collected and directed to the source server that can best serve the request using load balancing and application delivery controllers.

These features, along with others such as static objects and content caching, are already served in some areas, such as content delivery networks (
CDN), these services act as an aggregation layer for customers.
although
CDN offers more functionality in its edge point of presence (POP) than ever before, but edge computing guarantees the ability to run its own custom stateful applications in a protected space. More advanced fringe services include:

Accelerate the delivery of dynamic content
Retrieve data and perform fast delivery of personalized user experiences using enterprise application logic in Edge Pop

compatibility
Separate data to specific areas and perform functions in place, rather than transferring data to incompatible areas.

Perform data collection and analysis closer to the data source for time-sensitive applications.

An application field of edge computing: video surveillance

For example, an enterprise use case could include high-definition video surveillance of corporate offices and manufacturing sites. The purpose of using hd video is to improve the speed and accuracy of identifying unauthorized entry into buildings or sensitive areas. Applying AI and machine learning to video sources that companies want to use will lead to significant advances. However, moving large amounts of video data to a remote, centralized cloud data center to perform these computationally intensive operations is expensive. This problem multiplies as the number of sites that need to be monitored increases.

One solution is to be near the camera itself
(even in some cases on campus, or even on equipment) for basic processing. Tagged video clips can be passed to an aggregation edge, which has available computing and storage resources that can compare images to various sources and then generate valuable insights in the form of improved object or human detection.

Implementing a software control network allows an architecture that enables an enterprise to learn from
Benefit from better performance and security at the WAN edge. The strategy also allows for future expansion of the overlay network to access and aggregate edge POP in carrier-neutral data centers. This means that as end users’ demands for high performance, security and personalized experiences continue to grow, IT managers provide enterprises with maximum flexibility in computing and networking requirements.

Mark Casey, CEO of Apcela

Translation: ‘pala

Said by: Edge computing community edgew.net