Edge computing is a trendy topic nowadays. With the increasing amount of data that needs to be processed and stored, problems with infrastructure management, data security, privacy, and governance needs to be taken into account now more than ever.
According to a research report from MarketsandMarkets, the edge computing market size is expected to grow to USD 15.7 billion by 2025. It seems promising, right?
But what is edge computing?
"Edge computing is the practice of processing and storing data either where it’s created or close to where it’s generated — “the edge” — whether that’s a smartphone, an internet-connected machine in a factory or a car." [2]. The ultimate goal of edge computing is to reduce latency or the time it takes for an application to run or to execute a command.
What kind of application can benefit edge computing?
Typical applications for edge computing are those that generate huge amounts of data and that data needs to be processed as close to real time as possible, such as self-driving cars, augmented reality apps, and wearable devices.
The growth of the Internet of Things (IoT) applications across industries might be the key factor driving the success of edge computing, since IoT requires low-latency processing and real-time, automation of decision making, and needs an increasing amount of data and network traffic.
How to implement edge computing?
From the development perspective, there are two ways to approach edge computing, namely serverless and containers.
Containerization technologies, such as Kubernetes and Docker, allow code portability (i.e. deploy the code to different locations). However, serverless can be used for applications in which it is necessary to stream content close to the user. Typically, serverless is used for running applications in the cloud and containers for transforming hardware resources into the private cloud [1].
Now that you know what edge computing is, let's look at some edge computing trends.
1. Machine learning at the edge
Machine learning at the edge can be challenging due to the complexity of data solutions. However, nowadays there are AI-optimized hardware, container-package applications, and frameworks such as Tensorflow light and TinyML that makes machine learning at the edge more feasible. Additionally, open standards such as the Open Neutral Network Exchange (ONNX) are pushing the limits of machine learning interoperability by making on-device machine learning and data analytics possible on the edge.
2. 5G networks pushing edge computing
With the adoption of 5G networks, edge computing will be more feasible in the context of augmented and virtual reality and autonomous vehicles
3. Data security is improve with edge computing
With edge computing, data security is improved since the application does not depend on a single point of storage. With that, the user experience is improved as well.
4. Cloud and edge providers explore partnerships
According to the IDC Research, a quarter of organizations will improve business agility by integrating edge data with applications built on cloud platforms by 2024. Although that will require partnership between cloud and communication service providers (such as wireless carriers and major public cloud providers), applications closer to end users and with real-world assets will become an increasingly critical component of this trend.
About the author:
Isabella Ferreira is an Ambassador at TARS Foundation, a cloud-native open-source microservice foundation under the Linux Foundation.
References
[1] https://www.eweek.com/cloud/why-the-future-of-computing-is-at-the-edge/
[2] https://builtin.com/cloud-computing/future-edge-computing
