What are best Edge Computing Platforms?

Adlink Edge IoT Affirmed Cloud Edge Alef Edge AWS Edge ClearBlade Dell Eclipse ioFog Eurotech Edge Software EdgeConneX Equinix Edge Google Distributed Cloud Edge HPE Intelligent Edge IBM Edge Computing Infiot ZETO Intel Edge Technology and Solutions Microsoft Azure Stack Edge and IoT Edge Mutable Public Edge Cloud Schneider Electric Edge Computing Solutions Section.io Edge Compute Platform Vapor IO Kinetic Grid

Edge Computing: Essential Facts & Top 20 Solution Providers

Q: Why do we need Edge Computing when we have Cloud Computing?

Organizations can use a mix of both Cloud Computing and On-Premise Datacenters to solve their problems. They can balance their privacy, security, and budget needs by choosing between these two options as they deem fit in each usecase. There are plenty of product and service providers for both cloud computing and on-premise data centers. A good competitive market ensures that costs are reasonable and good quality is easy to find. The deployment of on-premise computing solutions is local, while Cloud Computing resources could be located anywhere in the world, and are offered via the Internet. Their commonality is that we use them for medium, large, and huge scale solutions. They have both evolved to solve the needs of organizations upto 2010s. There is some uniformity and standardization in place, both technically and in terms of expectations in the minds of professionals.

August 17, 2022

Home Automation, Innovation, Softwares

Edge Computing is a distributed computing framework choosen for improving response times, reducing bandwidth requirements, and processing data closer to its creation and consumption. Many new usecases and scenarios are emerging thanks to new developments like IoT, AI, and 5G where using cloud computing solutions is not enough.

In this article, we will understand the concept of Edge Computing, its various usecases, and the top providers of edge computing solutions. We will see how Edge Computing is distinct from traditional on-premise computing and cloud computing. What are the essential needs that require the application of Edge Computing?

What is Edge Computing? How is it different from Cloud Computing and Traditional Computing?

Let us understand the concept of Edge Computing logically with simple explanations of related concepts and how they fit together.

Network Edge

In the fields of Computer Networks and Distributed Computing, there is a term we call as "Edge of the Network" or "Network Edge". We use it to talk about the area in the network where a local device or local network connects with the larger Internet. It is the set of entry points for the Internet into the local network. And by corollary, it is the set of exit points for all devices located inside the local network.

The network edge represents a crucial boundary for the local administrators. This is the edge or border of the computer network that they can control directly. So, the local network admins can do the security, maintenance, and monitoring only for the devices within this boundary. Anything beyond this is like wild west that is under the rule, control or management of others.

Edge Devices

Any kind of devices that we place at the network edge are "Edge Devices". Typically, these are networking devices like routers, modems, routing switches, multiplexers, and integrated access devices. We can also include any Metropolitan Area Network (MAN) and Wide Area Network (WAN) access devices. However, the scope of edge devices is not limited to simply networking devices.

We can also keep any kind of computing, storage, and electronic device on the edge of the network. This will help all other local devices to share and utilize the computing, data storage, and other services that these devices can offer. Thus, it can be a cost-effective and localized solution to improve the capabilities of our entire local network, in an organization, institution or company.

However, this resource-sharing aspect is true for any device in the network, not just for edge devices. There is a crucial distinction between any local device and an edge device. A local device need not be connected to the Internet. It can be set to receive signals and commands only from within the local network. Whereas an edge device is expected to interface with both the local network as well as the Internet (or the Cloud or other remotely located devices and networks).

A Brief Look at Computing Paradigms

Let us understand Edge Computing by looking at the evolution of different computing paradigms.

Early Computing involved bulky machines that filled entire rooms. Computer monitors or stations called terminals were simply a way to access and use these room-sized computers.

Gradually over time, transistors, the building blocks of modern computers, underwent miniaturization. It became possible to bundle together computing, storage, and networking capabilities into desk-sized machines. This is how Personal Computing started trending, and became ubiquitous in most offices and homes.

Larger organizations have larger needs that Personal Computing devices alone cannot fulfill. This led to solutions like On-Premise Data Centers and Local Servers in offices. We can call this Traditional Computing (to contrast it with Cloud Computing).

With Apple iPhone and Android, we got mobile phones with computing capabilities. Smartphones and tablets turned the domain of cellphones with basic mobile telephony into Mobile Computing. This grew even faster than Personal Computing revolution, as smartphones are much cheaper and therefore affordable to a larger market. This brought computing to almost every pocket and at everyone's fingertips.

The Case for Cloud Computing

Even today, most organizations still prefer Traditional Computing solutions like local servers, local networks, and local data centers. This helps them secure their trade secrets, proprietary algorithms, confidential data etc. However, it has the downsides of maintenance costs. Scalability is slow and costly, both while upscaling and downscaling. Having dedicated staff to look after infrastructure has its overheads.

Parallel to these disadvantages of Traditional Computing, we also got the phenomenon of giant tech companies like Amazon, Google, Apple, Microsoft, and Facebook. Since they had to operate at such a large scale, they had to find solutions that optimized their spending and computing needs. They gradually became really good at taking care of their own computing needs with their custom-built machines and datacenters.

Economies of scale gave them the advantage that it was relatively much cheaper for them to add more computational, storage, and networking power. A new revenue stream became available by simply scaling up these operations further, and offering these computational resources on lease to anyone via the Internet, or the Cloud.

And this solution of leasing compute power as required proved to be a cost-effective and flexible option to other smaller companies. Cloud Computing is a boon especially for startups, who can simply focus their budget and efforts on their product and code, and leave the infrastructure worries to their cloud providers.

Why do we need Edge Computing when we have Cloud Computing?

Organizations can use a mix of both Cloud Computing and On-Premise Datacenters to solve their problems. They can balance their privacy, security, and budget needs by choosing between these two options as they deem fit in each usecase. There are plenty of product and service providers for both cloud computing and on-premise data centers. A good competitive market ensures that costs are reasonable and good quality is easy to find.

The deployment of on-premise computing solutions is local, while Cloud Computing resources could be located anywhere in the world, and are offered via the Internet. Their commonality is that we use them for medium, large, and huge scale solutions. They have both evolved to solve the needs of organizations upto 2010s. There is some uniformity and standardization in place, both technically and in terms of expectations in the minds of professionals.

But both these computing paradigms are rendered inadequate by advances like Internet of Things (IoT) and Artificial Intelligence (AI). The quantity and variety of consumer and industrial products that have aspects of IoT and AI is growing exponentially. And for a number of reasons that we will see next, applying On-premise or Cloud solutions to IoT and AI often isn't a viable choice. This is where Edge Computing comes in. It does not replace either Traditional On-premise Computing or Cloud Computing.

Edge Computing exists to do the kind of varied and unique tasks that neither of the two can do, because they were never designed to do that. It is an additional tool in your arsenal as you execute your projects. Sometimes, Edge Computing is also called Fog Computing, because it is closer to the "ground" like fog, while clouds are something far away and distant.

Edge Computing is a new paradigm in distributed computing which seeks to decrease the dependency on Cloud Computing. And while it is local, it is distinct from On-premise Data Centers because it involves new and unique usecases, often at tiny, small or medium scales.

The Top 5 Essential Needs that Edge Computing Solves

Looking at it in general terms, these are the top 5 needs why anyone will choose Edge Computing in their projects:

Real-time, Low Latency data processing needs. Immediate solutions are required.
High Volume of Raw Data such that it is impractical to keep sending so much raw data to the Cloud.
Poor or Spotty Internet Connectivity; for a number of reasons, whether remote location, congested urban area, or a high-level security facility facing DDOS attacks, etc.
Keeping Data and Processing Local: Data cannot be allowed to live on Cloud, and has to be kept on premise, whether for privacy, security, regulation or compliance reasons.
Seamless: Edge Devices can seamlessly connect to both local networks and the Internet. So data processing is predominantly local, while maintenance, monitoring, updates and management of edge computing facilities can be done remotely.

Thus, Edge Computing bridges the gaps between Traditional and Cloud Computing, and solves many new problems that cannot be tackled using old paradigms.

Why will Edge Computing be a Top Tech Trend in the 2020s and beyond?

Consider these estimates for the state of our world due to Internet of Things,

The number of IoT devices is predicted to be anywhere between 30-75 Billion in 2025.
Together, they will generate almost 80 zettabytes of data in 2025 (one zettabyte is a trillion gigabytes).

While Edge Computing cannot entirely replace Cloud Computing, it will take on a increasingly larger share in the market. This is because the need for "intelligence" at a local, device-level is increasing as innumerable new applications are being found for AI and IoT tech. These devices collect a lot of data and require real-time feedback to keep operating safely and accurately. It is infeasible to relay all that raw data to the cloud, get it analyzed, and then wait to receive the decision signals back. Both the volume of raw data being generated, and the requirement for low-latency, real-time feedback, make the case for Edge Computing over Cloud Computing.

Think of Edge Computing as a thin intelligence layer that can take many forms, and can be spread throughout the planet. It is geographically located much closer to the devices and local networks where the action is happening, unlike Cloud Computing. Edge Computing might be on-device, such as a self-driving car with on-board AI algorithms processing the high volumes of data coming from various sensors and cameras. Or its deployment can happen on the edge of the network, such as a micro data-center to communicate with the network of sensors and actuators on a farm, greenhouse, or research facility.

What are some Popular Applications and Usecases where Edge Computing is Essential?

The use of sensors and real-time AI processing will keep rising due to these kind of scenarios:

Smart Cities with intelligent surveillance, traffic management, preemptive resource allocation, and timely response in critical situations.
Self-Driving Cars that cannot afford to wait for Cloud response for critical real-time feedback.
Retail Stores and Hotels with Self-Checkouts and No Checkouts.
Vertical and Indoor Farms with Hydroponics & Aeroponics.
CCTV systems that want to increase efficiency by using motion-detection and other AI methods to filter only the abnormal or interesting events.
Home Automation and Personal Assistants like Nest Thermostat, Smart Speakers, Smart Doorbells, etc.
Autonomous Delivery Drones and Terrestrial Robots moving through traffic and/or areas with low cellular/Internet connectivity.
Industrial IoT applications in factories, manufacturing facilities, warehouses, supply chain management, etc.
Patient Monitoring Systems in hospitals that need to collect and process data from various devices and sensors, and build a consolidated, updated report.
Content Caching and CDNs are actually one of the earliest examples of Edge-like solutions. Streaming services like Netflix, Spotify, Pandora use these geographical caching techniques that are closer to the users, reducing latency and improving their quality of service.
Even as 5G, 6G and next gen cellular networks proliferate, offering higher bandwidths and lower latencies, they only inspire more Edge Computing usage. This is because as developers and creators envision more ambitious projects that take advantage of 5G bandwidths and speeds, they quickly overwhelm and exceed the capacities of cellular networks and clouds. Physics itself limits us ultimately, as the speed of light is 3,00,000 km/s. We can never transmit data faster than that. Hence, local, on-the-network-edge deployment of intelligence. processing, and storage will keep Edge Computing trending.

This is why, we will see the deployment of Edge Computing more and more, and it will interface with both Traditional and Cloud Computing services as well, to provide a high quality of service.

Summary

Edge Computing is an emerging distributed computing paradigm, that is expected to take a larger market share than Cloud Computing. The importance of edge computing is growing with the rising applications of Internet of Things (IoT) and Artificial Intelligence (AI) across smart cities, Industry 4.0, enterprise-level and end-consumer solutions. The primary idea is to reduce latency and ensure high quality of service by bringing computation, storage, networking, and intelligence closer to devices and local networks, rather than depending only on Cloud Computing. Edge Computing is essential to manage huge data volumes and real-time autonomous decision making that will become more and more common in the 21st century.

Devavrat Ravetkar

Edge Computing: Essential Facts & Top 20 Solution Providers

What is Edge Computing? How is it different from Cloud Computing and Traditional Computing?

Network Edge

Edge Devices

A Brief Look at Computing Paradigms

The Case for Cloud Computing

Why do we need Edge Computing when we have Cloud Computing?

The Top 5 Essential Needs that Edge Computing Solves

Why will Edge Computing be a Top Tech Trend in the 2020s and beyond?

What are some Popular Applications and Usecases where Edge Computing is Essential?

Top 20 Edge Computing Platforms and Solution Providers

1. ADLINK Edge IoT

2. Affirmed Cloud Edge

3. Alef Edge

4. AWS Edge

5. ClearBlade

6. Dell Edge Computing

7. Eclipse ioFog

8. Eurotech Edge Software

9. EdgeConneX

10. Equinix Edge

11. Google Distributed Cloud

12. HPE Intelligent Edge

13. IBM Edge Computing

14. Infiot ZETO

15. Intel Edge Technology and Solutions

16. Microsoft Azure Stack Edge and IoT Edge

17. Mutable Public Edge Cloud

18. Schneider Electric Edge Computing Solutions

19. Section.io Edge Compute Platform

20. Vapor IO Kinetic Grid

Summary