Edge computing and data security

Do you want to prevent your data from being found at a competitor's house or being exposed in the public square? We explain the risks associated with Edge Computing and the solutions to them.

 

Edge computing (or IT on the periphery) is on the rise in companies. They save valuable time by processing data locally at a time when the amount of data created daily is dizzying. Nevertheless, Edge Computing's solutions raise important information security issues.

 

security-camera-1300 px

What is edge computing?

While most IT specialists have no difficulty understanding this concept, it is different from employees in other departments. Good prevention of cyberattacks always starts with a good understanding of the risks and weaknesses at all levels of the company. In any case, a small refresh is always useful.

The rise of edge computing is going hand in hand with the rise of data and connected objects. According to the latest estimates from the International Data Corporation, there will be nearly 42 billion connected objects on the planet by 2025. Each of these objects will be able to produce data at all times. The amount of data created daily will be equivalent to 463 exabytes. Regardless of the scale of comparison used, such quantities are difficult for the human mind to grasp.

What should be remembered in the context of edge computing is that this mass of data saturates bandwidth and creates unacceptable latency.

Therefore, solutions are put in place to carry out part of the processing of data locally through a computer, server, or other equipment. These devices represent a kind of micro data center. The raw data is not sent to the cloud as it is, but is first pre-processed. Only relevant information, especially in the context of Big Data, is communicated to centralized data centers.

Edge Computing's solutions are preferred when near-instant data processing is essential. The classic example is the connected car. If you're parking and your car proximity alarm beeps when you're in the bumper of the nearby car, the added value is somewhat reduced.  At the enterprise level, real-time data processing is often essential to achieve the desired goal. One can think of a trading algorithm on the financial markets, a medical monitor or even a system to monitor a production line. Today, edge computing is an integral part of an effective data processing system.

Edge computing and cybersecurity

Connected objects and Edge Computing also means more potential gateways for computer attacks. Hackers are spoilt for choice when choosing the least secure gateway. If you want to take a fortress, you can attack the drawbridge with the full arsenal, look for a back door or attempt a breakthrough through a pipe or sewer. The front door will be different but the result will be the same: you will be in the stronghold. It's exactly the same when we talk about Edge Computing.

The risk is real. According to the University of Maryland, hackers attack an average of 2,244 times a day. More worrying is the data collected by IBM, which estimates that a security breach is usually discovered only after an average of 206 days. Beyond data protection, the financial cost is also substantial. Gartner estimates that annual overall cybersecurity spending in 2022 will be $134 billion.

The main difficulty of cybersecurity on the periphery is that it requires attention at different levels. Indeed, Edge Computing operates on the model of a distributed network. As a result, vulnerabilities are multiplied.

data-screen-13000 px

Conceptual vulnerabilities and physical vulnerabilities

Edge Computing relies on connected objects and micro data centers. Connected objects are often the weakest link and the most obvious gateways for hackers. It is often enough to access an object to pollute the entire network.

Edge computing equipment is also vulnerable. Instead of having to attack a gigantic data center, hackers have a multitude of options. The infiltration of these micro data centers can be done through hardware (by manipulating or simply stealing it) but also through classic attacks type DDoS or by the infiltration of the software used. Vulnerabilities therefore have very different faces.

Currently, the "security by design" approach is encouraged to minimize risk. As a reminder, this approach means that connected objects or micro data centers are protected from attacks from their creation. This may seem obvious, but in practice we find that too many companies are not allocating enough time and money to this preliminary stage, which will avoid many headaches.

In any case, even if more and more companies rely on the "security by design" approach, they cannot be satisfied with it. Aggression techniques are evolving and it is not excluded that a perfectly protected object today may be attacked tomorrow. So, relying only on what you've learned is not an option when it comes to security and Edge Computing.

Edge computing a cybersecurity challenge

Paradoxically, Edge Computing takes us a step backwards in the way we operate. While the last few years have been entirely dedicated to the cloud, the return of data and equipment to the company implies a proactive and dynamic approach. By 2025, 75% of the data will no longer be processed in centralized data centers. As you will have understood, despite the potential security risks, the benefits of Edge Computing will make this type of network unavoidable in the future.

Beyond physical security, it is the network architecture as a whole that needs to be analysed. What data can be processed with Edge Computing? What data should be sent to the data center? What are the connected objects? What software should be used? What is the sensitive data? All these questions involve deep and effective thinking in order to put in place a system that suits your needs and your business.

At Ryax, we offer a global Smart Data solution to process data in the right place at the right time while ensuring its security. If you would like to know more or better know our products, please contact us now.

The Ryax Team.