Jira Cloud is undergoing a major shift to become a cloud-native, multi-tenant platform designed for better scalability, speed, and reliability. The previous architecture, which combined cloud-specific layers with a single-tenant database, limited performance and made it difficult to adapt to the needs of modern users. The original design, rooted in early 2000s technology, relied on a relational database model that didn’t align well with Jira’s read-heavy usage. With a 10:1 read-to-write ratio, the old model created inefficiencies, especially in a multi-tenant environment. To improve performance and reliability, Atlassian is rewriting substantial portions of the code and rethinking how data is accessed and managed. They are transitioning from a monolithic architecture to a more modular system with dedicated services for different functions. For example, the Jira Issue Service (JIS) will manage all issue data, while the Jira Scalable Issue Search (JSIS) will enhance search capabilities. This shift allows for horizontal scaling and better resource management, enabling faster response times even under high load. The goal is to meet the performance expectations of large enterprise customers, aiming for an internal Service Level Objective (SLO) of at least 99.99% reliability. The new architecture not only enhances speed and efficiency but also aims to boost developer productivity by simplifying the codebase and reducing the need for custom fixes. The focus on service sharding and multi-level caching further ensures that the platform remains resilient and responsive, even during traffic surges.