August 2022 Summaries

Redis is investing in its Terraform registry, introducing new updates to version 1.0.0 for managing databases across multiple modules easily. The key changes include extracting the database resource from the subscription resource and introducing a new resource called 'rediscloud_subscription_database.' This allows users to manage their databases on separate logical resources and even by different configuration files if required, improving user experience and security. Additionally, users can now create or modify a database with multiple modules. However, there are breaking changes in the new version, so existing users must import their current configuration file to a new directory before upgrading.

"8 Data Modeling Patterns in Redis," a comprehensive e-book on data modeling in NoSQL, explores eight data models that developers can use to build modern applications without the limitations of traditional relational databases. Unlike relational databases, NoSQL allows for flexible schema designs, enabling developers to model data without prefiguring future changes. This approach simplifies data storage and retrieval, reducing overhead and improving performance. The e-book introduces various data modeling patterns, including the Embedded Pattern, which bundles two tables into one, and the Aggregate Pattern, which precalculates fields during writes to reduce read time and overhead. It also covers the Bucket Pattern for storing time-series data, the Revision Pattern for tracking document changes, and the Tree and Graph Pattern for handling JOIN operations. The Schema Version Pattern allows developers to pivot easily between different data models, making it a valuable tool for building modern enterprise applications. By leveraging these patterns, developers can build scalable and flexible applications that adapt to changing business needs.

The Redis OM Spring project has made significant progress, introducing new features that simplify and optimize Redis-powered Spring applications. The Redis Stack distribution, released in March 2022, integrates several popular Redis modules, including a JSON document database, search engine, time-series database, graph database, and probabilistic data structures. The Redis OM Spring client library helps model domain data and persist it to Redis in Spring applications, providing multi-model persistence of Java entities such as Redis Hashes and JSON Documents. New features include support for indexed collections, auditing metadata, and a Suggestions/Autocomplete API, making it easier to search and retrieve data efficiently. Additionally, the RedisBloom module provides probabilistic data structures to verify the existence of items in large datasets, improving performance and responsiveness. The project continues to evolve, with future releases focusing on the remaining modules and APIs.

Redis offers several probabilistic data structures - Bloom filters, cuckoo filters, Count-Min Sketches, Top-K data structures, and HyperLogLogs - to quickly process large datasets with some loss of precision. These data structures are used in various applications such as Google Chrome's URL filtering, IP address tracking, and network traffic monitoring. They provide fast lookups, efficient storage, and can be used to estimate the frequency of items in a dataset. The choice of which probabilistic data structure to use depends on the specific requirements of the application, including the desired level of precision and the trade-off between accuracy and space usage.

The Redis Insiders program was launched in May, introducing four ambassadors who will work closely with the Redis Developer Relations team to educate and enthuse the large user base about using Redis to solve problems. The ambassadors, Jyotsna from India, Moiz from France, Michael from Nigeria, and Stevan from Trinidad and Tobago, are passionate about communities and eager to share their knowledge of Redis through speaking, writing, and volunteering. They will act as the developer community's voice, providing feedback on pain points and suggesting improvements, and spreading the word about Redis via various channels. The program aims to make it frictionless for developers to get started with Redis and adopt it as a primary database.

Redis Enterprise is being adopted by leading companies such as Gojek, an Indonesian ride-hailing service, to power their machine learning (ML) platforms. Gojek has upgraded from Redis OSS to Redis Enterprise on Google Cloud due to its high performance, ease of management, linear scalability, and five-nines availability. The company's ML platform is now powering many of its use cases, contributing to its success as a "super app" valued at over $10B. The open source feature store Feast is also widely adopted by companies such as Better.com, Robinhood, and Udaan, and has become the most popular open source feature store since its launch in 2019. A quickstart tutorial on running Redis Enterprise with Feast on Google Cloud Platform (GCP) is available to help developers get started with this powerful combination.

The text discusses Redis, which initially started as a caching database but has evolved into a primary database. Many applications today use Redis as a primary database, although most service providers only support it as a cache. Using separate databases and caches can add complexity, compromise latency, and prevent the full potential of Redis from being realized. However, with Redis Enterprise, users can leverage both an in-memory cache and a primary database in one system, eliminating the need for multiple systems. This also allows for building modern applications and low-latency microservice-based architectures on top of Redis. Additionally, Redis Enterprise offers auto-scaling, clustering, and Active-Active Geo-Distribution features to ensure high reliability and global scalability.

Azure Cache for Redis Enterprise has introduced native JSON support, enabling developers to create a document database using the Azure Cloud Service Provider. This feature allows users to build modern applications with high-performance and scalability, supporting real-time use cases with low latency. With RedisJSON, users can ingest, index, and query JSON documents using native APIs, and perform full-text searches and complex structured queries. The addition of RedisJSON makes Azure Cache for Redis Enterprise the first Cloud Service Provider to offer a fully-licensed document capability as a first-party service, allowing seamless migration from other services. To utilize this feature, users can create a database with RedisJSON using the Azure Portal or AZ CLI command, and leverage tools like RedisInsight for data visualization and administration.

Redis OM .NET has released several new features since its initial version in November 2021, including indexing fields within embedded objects, arrays of strings within JSON objects, and support for indexing Enums, ULIDs, GUIDs, and Booleans. The library also allows developers to index and query aggregations, load specific or multiple fields from RedisAggregationResults, and provides a new API for updating and deleting items stored in Redis. Additionally, Redis OM .NET now supports configurable chunk sizes for automatic pagination under the hood, allowing users to customize the performance of their queries. The library is actively maintained and updated, with a GitHub Repository where developers can track new features and submit requests.

Redis OM for Node.js has made significant progress in its early days, with notable additions including a fluent interface for searching, improved support for expiring entities and a more seamless integration with Node Redis. The developer community's feedback and contributions have been instrumental in shaping the project, with many users providing useful suggestions and fixes. Key changes include adopting initial values for `createEntity`, adding support for expiring entities, and changing the default data structure from Hashes to JSON. Future releases are planned, including additional features and helpful content for new users. The project's success is largely due to community involvement and assistance.

Real-time applications require real-time solutions to meet customer expectations for speed. Redis Enterprise is an in-memory database that provides the speed and performance necessary for real-time use cases compared to traditional SQL databases. Caching user session data, APIs, and other parts of the data lake into Redis Enterprise can transform your data layer and enable real-time applications. Data ingestion methods like blue-green deployments, multi-phase migrations, legacy systems integration, hybrid cloud architectures, geo-distributed on-demand infrastructure, and CQRS pattern implementation using Redis Connect help maintain consistency between legacy and new databases while enabling real-time data use cases.