Skip to content

overview.md

Latest commit

 

History

History
30 lines (24 loc) · 2.39 KB

overview.md

File metadata and controls

30 lines (24 loc) · 2.39 KB

Redis Data Integration (RDI) is a powerful tool designed to help Redis Enterprise users sync their fast Redis databases with live data from slower, disk-based databases. The primary goal of RDI is to enhance the speed and scalability of read queries, ensuring a smooth and predictable user experience, especially as user demand grows.

Key Benefits of RDI:

  • Improved Performance: RDI caches data from read queries, enabling your application to handle millions of users without major redesigns.
  • Cost Efficiency: By reducing the need for expensive database read replicas, RDI helps lower the total cost of ownership.
  • Simplified Operations: RDI eliminates the need for manual pipeline building and coding data transformations, saving time and resources.

How RDI Works:

RDI uses a Change Data Capture (CDC) mechanism to keep the Redis cache updated with changes from the primary database. It also allows you to transform data from relational tables into Redis-compatible data structures using a configuration system, eliminating the need for coding.

Key Features:

  • Near Real-Time Updates: RDI captures and processes changes in micro-batches, ensuring the Redis cache is almost always up to date.
  • Data Integrity: RDI maintains the order of data changes, ensuring consistency across your database.
  • High Availability: With hot failover and quick recovery, RDI ensures minimal downtime.
  • No Coding Required: The system is easy to install, operate, and configure using Redis Insight and a command-line interface (CLI).
  • Security and Observability: RDI encrypts data in transit and offers robust observability through metrics, logs, and Prometheus endpoints.
  • High Throughput: RDI can process around 10,000 records per second, making it suitable for high-demand applications.

When to Use RDI:

  • Your app relies on a slow database as the system of record.
  • You already plan to use Redis as a cache.
  • Data changes frequently and can tolerate eventual consistency.

When Not to Use RDI:

  • You are migrating data into Redis only once.
  • Your data is updated infrequently in large batches.
  • Your app requires immediate cache consistency.
  • The data set is small or you need to write data to the Redis cache first.

RDI is a solution tailored for applications that need the reliability of a disk-based database but also require the speed and scalability of Redis to handle growing user demand.