Google this week announced that it will begin the first public testing of a new platform that uses machine learning to detect and prevent denial of service (DDoS) attacks. Offered to enterprise customers, Google Cloud Armor Adaptive Protection also promises protection against HTTP overloads and other layers of other frequency.
According to the description offered by the company, its artificial intelligence is capable of detecting and alerting on anomalous activities, generating signatures describing the possible attack and creating a custom WAF (Web Application Firewall) rule to deal with the situation. The technology is the same the company uses to protect its own systems, and which has helped it deter massive attacks in the past.
Producer Manager Emil Kiner explains that the problem of DDoS attacks has grown exponentially with the creation of new tools and the emergence of botnets that can be rented by attackers. “Since attacks can come from millions of individual IP addresses, manual screening and analysis to generate and apply blocking rules is time-consuming and resource-consuming, allowing high-volume attacks to affect applications,” he explains.
Technology available to everyone for a limited time
Machine learning-based technology is capable of detecting high intensity DDoS attacks against internet applications and services, reducing their impacts and ensuring information traffic. The company claims that adaptive protection can be configured based on each company’s security policy, and that the feature is now available to all Google Cloud subscribers.
However, Google notes that this should only last as long as Google Cloud Armor’s adaptive protection is in the period it calls “Preview”. “Once Adaptive Protection is generally available, it will only be available to customers who subscribe to Cloud Armor Protection Plus,” he notes.