In late 2024, a Mirai-based botnet launched 5.6 terabits per second of traffic at an Asian ISP. It was the largest DDoS attack ever recorded and lasted just 80 seconds. And that’s all it took to make headlines and redefine the threat landscape. Attacks over 1 Tbps are no longer rare. In fact, one report showed a 1,885% quarter-on-quarter spike in events at this scale.

What’s changed isn’t just the volume, but the strategy. Attackers now favour short, high-impact “blitz”  floods during peak business hours. They pulse traffic in waves to evade detection or scatter it across vast IP ranges in a technique called “carpet-bombing.” These attacks can saturate upstream links and collapse networks before on-premise defences can react. Legacy scrubbing appliances were not built for this kind of speed or volume.

That’s why Horizon 1 exists. It’s the outermost layer of RedShield’s three-horizon DDoS and bot defence model and is built on AWS Global Accelerator and Shield Advanced. This layer provides always-on, global-scale DDoS mitigation. All traffic is routed to the nearest AWS edge location where terabit-scale filters strip out volumetric junk before it reaches your infrastructure. With packet-level inspection and live global threat intel, RedShield’s protection blocks known attack signatures, malformed requests, and risky traffic sources at the edge.

This model isn’t just theoretical. When one of our customers in New Zealand was targeted with a DDoS wave that surged from 50 Gbps to 2.3 Tbps in minutes, RedShield’s service absorbed the entire load at the edge. The customer stayed online, users saw no disruption, and the attack collapsed within the hour.

With 5G and fibre continuing to increase available bandwidth, attacks at this scale and beyond will only become more common. RedShield’s DDoS protection is already built for that future.

 

Learn more in our whitepaper:

Discover how RedShield neutralises terabit-scale DDoS attacks before they reach your infrastructure. Learn how we combine volumetric protection with deeper behavioural and identity-based controls. Contact us now.

 

Next article: Generative AI is Exposing the Limits of WAFs. RedShield’s Approach Addresses the Gaps.
All Knowledge base