Episode Details

This story was originally published on HackerNoon at: https://hackernoon.com/how-agoda-scaled-its-feature-store-50x-with-scylladb.
Learn how Agoda scaled its ScyllaDB-powered feature store 50x using NVMe upgrades, cache optimization, and smarter data modeling.
Check more stories related to tech-stories at: https://hackernoon.com/c/tech-stories. You can also check exclusive content about #agoda-scylladb-scaling, #scylladb-nvme, #feature-store-architecture, #cache-stampede-prevention, #dragonflydb-caching, #scylladb-compaction-strategy, #low-latency-database-scaling, #good-company, and more.

This story was written by: @scylladb. Learn more about this writer by checking @scylladb's about page, and for more stories, please visit hackernoon.com.

After Agoda’s feature store traffic surged 50x, the engineering team faced severe latency spikes and cache stampedes threatening production stability. By benchmarking ScyllaDB under cold-cache conditions, optimizing SSTable summaries, experimenting with compaction strategies, improving caching, and ultimately replacing SATA SSDs with NVMe drives, Agoda boosted database capacity by up to 60x while maintaining 10ms P99 latency at massive scale.