May 2024 Summaries

In May, Redis introduced several updates and enhancements. AWS Marketplace customers can now create Essentials subscriptions on their Redis Cloud account and pay via AWS Marketplace, with a $5 minimum monthly cost and a 14-day $500 free trial available. Additionally, Redis Cloud now supports active-active deployments with search and query functionalities and a robust document store with JSON. Users can also set the dataset size on new databases for better memory allocation. Lastly, SAML single sign-on now supports multiple username domains to accommodate organizations using multiple domains.

Microsoft's developer toolkit, Semantic Kernel, now integrates with Redis to allow developers to build high-performance AI applications. The integration enables the use of unstructured multimodal data and common LLM design patterns such as retrieval-augmented generation (RAG). By using Redis as the back end for Kernel Memory, developers can achieve high performance and reliability in their apps. Semantic Kernel is designed to be platform-neutral, allowing it to work with any language that can make HTTP requests. The integration of Kernel Memory with Redis makes managing the memory layer of the semantic computer more intuitive and flexible.

Large language models (LLMs) are being used to answer questions about corporate data by connecting them to specialized data through retrieval-augmented generation (RAG). RAG works by integrating a retrieval component into the generative process, allowing LLMs to access private or corporate data and provide more accurate responses. However, traditional RAG approaches focus exclusively on text, leaving out information-rich images or charts contained in slide decks or reports. A new multimodal RAG template is being introduced, which allows models to process and reason across both text and images, paving the way for more comprehensive and nuanced AI apps. This template uses Redis and OpenAI's combined text and vision model, GPT4-V, to index documents and summaries efficiently and reduce redundant work by storing responses from previously answered questions in a semantic cache. The new template enables devs to build sophisticated AI apps that understand and leverage diverse data types powered by a single backend technology—Redis.

In multi-tenant architecture, a single software instance serves multiple distinct user groups or "tenants," with each tenant's data securely isolated from others. Within Redis, this approach allows for efficient management of tenants while maintaining their data securely and separately. Key advantages include operational efficiency and cost-effectiveness due to maximum resource utilization without the need for additional physical infrastructure for each new tenant. This results in simpler and more scalable business operations. Redis provides a multi-tenant environment on-prem or in any cloud infrastructure controlled by the user, making it particularly valuable for developing in-house apps using microservices. By adopting a multi-tenant approach, users can build and test in parallel with significantly reduced effort. In contrast to multi-instance architecture, where a new software instance is installed for each tenant, Redis's multi-tenancy handles the server or infrastructure level, ensuring that each tenant's operations remain isolated and secure. This approach closely resembles a multi-instance setup but simplifies provisioning and launching new Redis services provided by the management layer. Redis offers software multi-tenancy where a single deployment efficiently supports hundreds of tenants with distinct Redis endpoints, ensuring complete isolation from others. Deploying Redis in data centers, private clouds, or virtual private clouds leverages the economic advantages of this multi-tenant architecture. The Redis architecture layers on multiple levels of abstraction to achieve multi-tenancy, high availability, linear scaling, and high throughput, among other capabilities. Key components include nodes (hardware foundations), shards (core instances managing a subset of the total dataset), databases (logical endpoints for tenant data), zero-latency proxies (multi-threaded proxies routing Redis operations from clients to the correct database shards), and cluster managers (managing the entire cluster lifecycle). Redis's multi-tenant solution brings several benefits, including reduced infrastructure costs, seamless scaling, fine-tuning for high availability, persistence, eviction, and data size, agility in dev, testing, and prod environments, and compatibility with major cloud platforms. The shared-nothing architecture separates data-path components from control and management components, offering significant advantages such as improved performance, availability, security, manageability, scalability, and resource allocation. Redis has been successfully deployed in numerous organizations, including large e-commerce companies and SaaS providers, to handle data for different departments or clients while maintaining data privacy and efficient resource utilization. Security measures are in place to protect against potential breaches, ensuring strong access controls, regular security audits, and encryption standards to secure data both in transit and at rest. Overall, Redis's multi-tenant architecture offers significant benefits, improving performance, scalability, and security while efficiently managing resource allocation and data isolation, ensuring operational continuity and high availability. This makes it a strong cost-effective solution for enterprises looking to manage data across various environments.

The Redis team has introduced a new brand identity and redesigned its website, consolidating resources under redis.io. The company has also released several updates for Redis Cloud, including use case-driven database creation, support for AWS Transit Gateway, and improved cloud API functionality. Additionally, the Redis Enterprise Software UI has been enhanced with features such as managing locked users, changing passwords at login, and choosing the database version during creation. Finally, a new feature has been added to Redis Insight, allowing users to load demo data with one click. These updates aim to improve user experience, scalability, and performance for Redis users.