April 2024 Summaries

Vector databases are used as memory for AI apps, particularly those powered by large language models (LLMs). Redis has gained traction as a high-performance vector database and is now being used in the Spring community to simplify the development of AI-powered apps. A Retrieval Augmented Generation (RAG) workflow integrates data with AI models, where data is loaded into a vector database, and then used to generate responses based on user queries. The RAG workflow has been implemented using Redis as the vector database and Spring AI in this example, which uses a dataset of beers to demonstrate the concept. The implementation includes a JSON reader to parse the dataset, a RagService class that implements the RAG workflow, and a RagController class that exposes the service as a HTTP endpoint. This project showcases how combining Redis' speed and ease of use with Spring AI's abstractions makes it easier for Java developers to build responsive AI applications.

The recent announcement that Redis has achieved the Google Cloud Ready designation for its CloudSQL and AlloyDB services signifies that Redis' solution has met certification guidelines and has been validated by Google after a collaborative effort with their engineering teams. This achievement enables enterprises to adopt modern cloud-native solutions tailored to their specific requirements, such as real-time transactions, chat/messaging, gaming leaderboards, healthcare claims processing, and media streaming. By integrating Redis Cloud with Google Cloud databases, organizations can unlock the potential for real-time experiences, leveraging lightning-fast performance and robust resilience provided by Redis. This integration empowers enterprises to deliver exceptional user experiences while harnessing the power of Google Cloud's database services for data storage and analytics.

To catch up on the latest Redis releases from March, we have a major milestone for Redis Cloud with Fixed Plans available in the Google Cloud marketplace starting at $5 a month. A new client library called Lettuce is also joining our officially supported libraries, offering a consistent experience across functionality and support. Additionally, a RedisInsight update brings new formatters for vector embeddings in Browser, enabling easier visualization and integration with other tools. We have more valuable updates coming in April, so stay tuned for the next episode of "What's New in Two".