Introduction

PostgreSQL and SQL are renowned for their ability to maintain data integrity and offer robust query capabilities. Among open-source database management systems, PostgreSQL stands out as a top choice. However, when working with extremely large datasets that require high levels of concurrent access, it can face limitations. To address these challenges, Citus provides an architectural solution that distributes data across multiple nodes. This approach enhances performance and optimizes hardware resource utilization.

Architecture

Citus is designed around PostgreSQL servers that form a Citus cluster, with each server equipped with the Citus extension along with other extensions. It leverages PostgreSQL's extension APIs in two significant ways to modify the database's behavior:

• Replication of various database elements across all servers, encompassing custom types and functions.
• Introduction of two new table types, both optimized for increased scalability across multiple servers.
• Citus employs a technique known as sharding to achieve scalability. Sharding involves breaking down large databases into smaller chunks or shards and then distributing these shards across numerous nodes. This intelligent approach involves directing queries to the appropriate nodes and aggregating the results efficiently.

Key Factors

Citus has some important attributes:

• Horizontal Scaling: Unlike vertical scaling, which aims to enhance the performance of existing machines, Citus achieves scalability by adding more machines to the cluster.
• Parallel Query Processing: Citus leverages the aggregate query processing capabilities of all nodes, enabling it to execute queries in parallel across multiple nodes, significantly boosting performance.
• High Throughput: Designed for large-scale data applications, Citus efficiently processes vast amounts of data and queries, avoiding bottlenecks and maximizing resource utilization.
• Multi-Tenancy Support: It facilitates the development of applications with multiple tenants, allowing data to be distributed across various distributed tables.
• Familiar Compatibility: Citus's compatibility with PostgreSQL ensures a smoother learning curve, particularly for those already familiar with PostgreSQL, as it allows users to utilize familiar PostgreSQL tools, extensions, and methodologies.