Case Study: Akamai sharpens threat intelligence at 55% lower cost
Key results
The challenge
Akamai's Cloud Security Intelligence division ingests massive volumes of security event data, with its largest feed handling over three million events per second. It ran on Confluent Cloud with object storage, where Confluent Cloud had become one of its top third-party expenses, batch processing via object storage introduced high latency, and object storage had scaling issues on Linode.
The solution
Akamai adopted Redpanda Self-Managed Enterprise as a Kafka-compatible replacement, initially on Azure and then shifting workloads to its Linode cloud, while keeping existing DevOps tooling such as Terraform scripts and topic management unchanged.
“Confluent Cloud was one of our top third-party expenses. Redpanda came in at about 55% lower costs. It gave us the headroom to grow and opened the door to rethinking our whole pipeline.”
IHIdan HarelInfrastructure Team Lead, Akamai
The results, in context
Akamai reports Redpanda came in at about 55% lower cost than Confluent Cloud and self-hosted alternatives, replaced object storage limitations with a cleaner architecture, and enabled near real-time processing across global data streams. Three major upgrades were performed with zero downtime despite running on Akamai's private Linode cloud.