Case Study: SecurityScorecard achieves 90% lower latency and $1M savings with ScyllaDB

How SecurityScorecard built a scalable and resilient security ratings platform with ScyllaDB

SecurityScorecard, the global leader in cybersecurity ratings, needed a more scalable and resilient way to power its security ratings platform as its customer base grew. Its existing architecture relied on Redis, Aurora, and Presto/HDFS, but Aurora and Presto latency spiked under heavy throughput, Redis could not scale beyond its largest instance, and HDFS slowed score updates for days.

SecurityScorecard moved to ScyllaDB Cloud and introduced a new scoring API as a data gateway, routing low-latency requests to ScyllaDB and historical or complex queries to Presto + S3. With ScyllaDB, the company achieved over 90% lower latency on key endpoints, 80% fewer production incidents, $1M in annual infrastructure savings, and a 30% faster data pipeline, while also improving scalability, maintainability, and immediate score updates.

SecurityScorecard

Nguyen Cao

Staff Software Engineer