Case Study: SAS Institute improves read performance and migrates adtech workloads seamlessly with ScyllaDB

SAS Institute: Changing All Four Tires While Driving an AdTech Engine at Full Speed

SAS Institute, a major analytics and customer intelligence software company, faced the challenge of migrating its high-performance AdTech data platform from DataStax Enterprise to ScyllaDB. The core challenge was executing this migration without any downtime or data loss while the system was actively processing billions of monthly ad-serving decisions, a process they likened to changing all four tires on a moving car. Their key concern was maintaining fast read performance to ensure quick page rendering for users.

The solution from ScyllaDB was a seamless drop-in replacement due to its Cassandra compatibility. SAS Institute implemented a sophisticated "aggressive lazy reading" migration strategy using an abstraction layer, which allowed the switch with zero code changes. The results were highly successful: the migration was completed within a tight two-week window with no downtime, no lost data, and no customer complaints. Furthermore, ScyllaDB delivered a significant performance improvement, reducing average read latencies to low single-digit milliseconds.

SAS Institute

David Blythe

SAS Institute