A global electronic lock company wanted to upgrade their identity access management (IAM) system to achieve global scale with no manual sharding. They wanted to move away from sharding and evolve the app’s architecture from monolithic to microservices. With no in-house site reliability team, they started evaluating database-as-a-service (DBaaS) products. Their requirements included:
- Data domiciling: The manufacturer has customers across Europe, Asia, and North America, and they needed to ensure European customer data remained in Europe in order to remain compliant with the General Data Protection Regulation (GDPR). If possible, they wanted to pin row-level data to individual countries.
- Global scale without manual sharding: The team wanted to scale globally to reach its widespread customer base, but they needed low-latency reads for their time-sensitive use cases.
- Resiliency: In the event of a regional failure, they needed their database to remain available so customers could access important facilities.
- Data log: log of their transactional updates into their Elasticsearch, Logstash, and Kibana stack as an event bus for microservices.
The team landed on using Managed CockroachDB, which is able to hit all four of the requirements through a managed, geo-partitioned global database with change data capture capabilities. Their new global setup has three regions, which not only improves latency for customers living in each region, but gives them the ability to domicile European data for GDPR compliance.
Managed CockroachDB is now a critical part of the application’s infrastructure, and the manufacturer’s small engineering team has more free time to handle development. Want more detail? Read the full case study here.
