Google is using machine learning to stop DDoS attacks
Google has revealed a preview of a new machine-learning program from Cloud Armor called Adaptive Protection. The machine-learning-powered tool detects and protects enterprise applications from DDoS attacks. The same technology has been implemented in Project Shield, a free service offered by Google that aims to protect human rights, government, and media organizations against harmful DDoS attacks. In the past, Google has been effective in blocking huge DDoS attacks, including one in 2017 that measured in at 2.56 Tbps. Google eventually traced the attack back to a Beijing-backed attacker.
The new project was announced in November as part of its DDoS defense and web application firewall service that allows Google customers to use the same technology the company uses to protect itself. The Adaptive Protection technology uses machine learning models to effectively analyze signals across web services, therefore detecting potential attacks before they occur. The program can detect high-volume application-layer DDoS attacks. It is also able to accelerate mitigation by uncovering abnormal traffic. The public preview allows Google Cloud customers to test out the functionality of the product.