It is a widely accepted fact that IT spectra have been dominated by Cloud computing for the last two decades. It is the advantages of shared data centers, infrastructural offloading, scalability in service and of course reduced cost that deliberate companies for cloud migration. However, the core focus of centralization may be over by now, and we see a sharp inclination towards edge computing.

Does that mean new opportunities lie with edge computing? Or is that the cloud computing key issues like latency, virtualization related problems? Or a number of server placement that opens up the new scope with edge computing? Probably it is just one of the reasons for edge computing 2019 focus.

A significant driver behind the use of edge computing is also the rapid surge of the Internet of Things technology. However, its key success is its architecture for intelligent applications which are mainly data driven. Interestingly, in industrial applications also, the use of IoT devices are growing high.

Initially, industrial organizations involved cloud computing, big data analytics and machine learning to get core insights from industrial data. However, at later point of time they switched this analytics capability to smaller devices. Instead they target to sit them near the sources of the data. This helps to get the more real-time report. This is where edge computing comes into the picture.

But what do we mean by edge computing and what is it doing in industry space?  What are the edge computing working principles? Let’s have some overview.

What is edge computing?

According to research firm IDC, edge computing is described as “mesh network of microdata centers that process or store critical data locally and push all received data to a central data center or cloud storage repository, in a footprint of less than 100 square feet,”

Hence, we can say the word ‘edge’ here signifies geographic distribution. Unlike cloud computing, in this case, computing is not performed at the data centers which may be geographically distributed and far away from the source of the data generation. Also, it places an additional tier between end services and the cloud.

However, if we trace back to the history the concept is not very new. There was a step by step development in this space.

– Step 1: Akamai initially launched the idea in the 1990s with CDN (Content delivery network) which introduced data nodes close to the data sources. These nodes were supposed to cache static data. Edge computing took one step ahead by adding computing capability with that.

–Step 2: In 1997, edge computing was demonstrated for the mobile speech recognition purpose by the computer scientist Brian Noble.

-Step 3: In 1999, the above method was used for extending the battery life of mobile phones. At the same time, peer-to-peer computing arrived in the market.

–Step 4: In 2006, with the emergence of Amazon’s EC2, it received immense popularity. However, in 2009 the latency is analyzed in cloud computing end to end relationships. Thus it brought up a new concept – cloudlets. This started shaping modern age edge computing.

–Step 5: In 2012, as Cisco introduced “Fog computing” regarding IoT, we entered into the current edge computing era.

What are Key takeaways

-Low latency

-Resilient

-Reduced bandwidth and associated low cost

-Reduced server resources and hence the reduced cost

Edge computing Lexicons you must know

Now, while dealing with edge computing, we must know some key terms associated with it.

-Edge devices: This can be any device that produces data. These could be IoT sensors, small industrial machines or other devices which produce or collects data from other sources.

-Edge: This signifies the point where data enters or exits the edge network. Now it varies as per the scenario. For example, for mobile edge computing, the edge is the cell phone or the telephone tower. In an industrial manufacturing plant, it could be a machine.

-Edge gateway: A gateway is a window which is expanded beyond the edge network.

-Fat client: Software which can process in edge devices.

-Edge computing equipment: Any internet accessible equipment can be outfitted as edge computing equipment. Sensors, machines, and other devices can be used as edge computing equipment.

What are Edge computing Working principles

Edge computing adds an additional tier between the Cloud and the end-devices. This layer works between data sources and cloud data center. What is edge network? In an edge computing network, the primary data ingestion point is the cluster of edge devices. These devices are responsible for computing, storage, and networking.

Here each ingestion point is analyzed based on a complex event processing engine that follows specific pre-defined rules and policies. Based on the analysis, data processing either takes place locally which is called “hot data processing”. Otherwise, it is sent to the public cloud, commonly known as “cold data processing.”

Edge computing examples

There is extensive use of edge computing today. It is a misconception that it is only related to IoT devices. Instead we can find the use in every sphere of businesses. Somewhere it is used for health care or mobile computing or even in oil sectors.

# IoT:

With enhanced use of IoT sensors, it has become a need to analyze the generated data on a real-time basis. No matter whether it is for autonomous vehicles or smart cities which generate a massive amount of data every day.

However, this is only possible if the data source and the analytics solution are sitting nearby. Otherwise, it may cause poor connectivity issue. Besides, it is not a good idea to constantly connect an IoT device to the centrally located cloud server.

# For processing latency-sensitive information:

Latency occurs when information traverse through long network path or multiple hubs. For financial transaction related services where a latency for milliseconds can cause issues, edge computing works as a tangible solution. As data does not need to traverse along for processing, it helps in reducing the latency.

# Oil mining:

In oil mining process the sensors attached to the machines generate a lot of data to ensure systems’ working performance. However, such data does not necessarily is required on a real-time basis; a preferably daily report is enough for monitoring. Hence, instead of sending such data over a network, configuring local edge device with the oil rig machine can solve the purpose. This, in turn, reduces the network load as well.

Related post – 5 Big data analytics trends for 2019 expected to influence Artificial Intelligence

# Mobile computing and 5G cellular networks:

With the rapid use of mobile and smartphone apps, increased memory consumption and demand for real-time data has become a real concern for the consumers as well for the network operators. Edge computing works as a solution here. Additionally, as per IDC predictions with introducing 5G into their wireless networks, telecom providers will increasingly add micro-data centers with edge computing 5g.

These data centers are supposed to integrate to the existing 5G towers of will sit adjacent to 5G towers. Furthermore, these data centers will be used for edge computing to get direct access to the telecom provider’s network.

# Healthcare industry:

Integrating IT into healthcare has been a challenging task for long. It is evident that IoT sensors have been introduced with healthcare monitoring devices. However, when it comes to the question of patient care, they fail to perform as expected. With the help of edge computing, such issue can be solved as data latency can be easily minimized. So, patients can access their patient data and also pathological support remotely with the introduction of edge computing.

Challenges associated with edge computing

Security challenges:

Due to its distributed architecture this type of network is more prone to malware attacks. Such attacks are commonly known as attack vectors which is nothing but to access edge computing network to inject malware programs.

Licensing challenges:

Edge clients are associated with licensing cost. Initially, though they can charge a low price, with enhanced functionalities, the cost may rise.

Configuration challenges:

Edge computing consists of complex networking architecture, and if not configured correctly, it may cause security loopholes. Hence, the configuration must be robust and centralized.

Final verdict

Over the next few years, it is expected that edge computing will expand rapidly with the skyrocketing use of IoT devices, mobile computing, autonomous vehicles, smart cities, etc. Also, the changing business models in every sector will continue in this growth.

Besides, it will help in changing existing technologies to shape in a more advanced way. For example, edge computing will broaden the improvement spaces in cloud computing as many backend services for edge computing will become the research areas  for cloud computing. Besides, few technologies like Fog computing, Blockchain are expected to mature more.

Please share your valuable inputs in comment area to make the article more informative.