December 2021 Summaries

The text discusses how Redis Enterprise can help scale applications for enterprise customers who demand low latency, fast failover with no data loss, and global distribution of their applications. Unlike other Redis providers that only offer caching services, Redis Enterprise provides features such as 99.999% uptime, sub-millisecond latency, single-digit-seconds failover, Active-Active Geo-Replication, and no data loss. It also offers multiple primaries spread across the globe for local, sub-millisecond latencies for both reads and writes, using Active-Active Geo-Replication based on CRDTs. Additionally, Redis Enterprise provides enterprise clustering, Redis on Flash, and other native functions that address specific enterprise solutions not found elsewhere.

In this post of the "Redis in 90 seconds" series, it is demonstrated how to use Redis with any major cloud providers or in a hybrid cloud setup. Unlike other Redis providers that host open source Redis and provide it as a cache without supporting it as a database, Redis Enterprise allows users to store all or most of their data in a single system, making it easy to move from one cloud to another. It is available on major cloud platforms like Amazon AWS, Google Cloud, Microsoft Azure, Heroku, and can be downloaded as software for hybrid deployment scenarios where private data is stored on-premises while the rest is stored in the cloud. This flexibility helps users avoid being "cloud locked-in".

The application is a real-time strategy game called Pizza Tribes that involves training mice to bake and sell pizzas for coins, with the objective being to generate more coins than any other player. The game uses Redis as its database, which enables efficient transmission of data between components, allowing for real-time gameplay without delays. The application was created using a combination of Go programming language, Docker containers, and websockets, with a focus on minimizing load on the Web API. The game state is accessed through different methods depending on the use case, including JSON.GET and ZREVRANGE commands on Redis. The update process involves a check-and-set approach to ensure that the user's record in the user_updates sorted set is updated correctly. The leaderboard is updated using ZADD command, and users can access it by running the ZREVRANGE command. The application has been successful in achieving real-time gameplay, allowing users from around the world to battle each other for the top spot.

Google's high standards for search engines have made users expect instant answers, leading to the development of Alexis, an application that uses Redis as its main database to provide fast and efficient search functionality. Alexis leverages RedisAI and RediSearch to retrieve information from a corpus in response to a query with maximum efficiency, allowing users to index PDFs and extract information with ease. The application uses NodeJS with TypeScript in the backend and React with TypeScript in the frontend, with a webserver exposed through the Express framework, including endpoints for user registration, login, logout, and querying documents. RedisInsight is used to visualize the data and optimize the Redis application. The application provides an efficient text ranker capable of retrieving search queries at maximum speed, allowing users to make relevant searches and receive instant answers.

This tutorial provides a step-by-step guide to using Feast with Redis as an online feature store for machine learning. It covers deploying a local feature store, building a training dataset, materializing feature values into the Redis online store, and reading the latest features from the Redis online store for inference. The tutorial demonstrates how to use Feast's Python library + optional CLI to install and configure the feature store, and how to use Redis as an online store to deliver real-time predictions at scale with low latency and high throughput.

Redis OM for Python is a new developer-centric library that provides object mapping, data validation, and other features. It allows users to model data with declarative models similar to SQLAlchemy, Peewee, and the Django ORM. The library also supports fluent query expressions and secondary indexes, as well as both asynchronous (asyncio) and synchronous programming in the same library. Redis OM for Python includes two base model classes: HashModel and JsonModel, which store data in Redis Hashes or natively as JSON objects, respectively. The library also generates globally unique primary keys using the ULID specification and provides Pydantic validation based on type hints in models. Additionally, it supports querying for data based on attributes other than the primary key through fluent query expressions built on top of RediSearch, a source-available module that gives Redis querying and indexing capabilities.

Redis OM is a new .NET client library for Redis that aims to provide a simple and intuitive way for developers to store and query domain objects in Redis. The library uses object mapping, secondary indexing, and simplified querying capabilities to make it easier to work with Redis. It works well with RedisJSON, which provides additional features such as document modeling and full-text search. With Redis OM, developers can easily connect to Redis, create indexes for their data, and build complex queries using LINQ-like syntax. The library also supports Redis aggregations, allowing developers to calculate summary statistics and group records together. Redis OM is designed to be easy to use and flexible, with a declarative model for defining indexes and a simple API for connecting to Redis.

Redis OM Spring` is a new client library designed to model domains and persist data to Redis in Spring applications, providing additional indexing and querying capabilities with `RedisJSON`. It aims to provide high-level abstractions idiomatically implemented for various languages and platforms. The library builds upon `Spring Data Redis`, offering robust repository and custom object-mapping abstractions, as well as declarative Data Repositories that utilize `RediSearch` to index JSON documents. With its ability to create repository implementations automatically at runtime from a repository interface, Redis OM Spring provides a compelling developer experience for building large applications with enterprise features.

This library makes it easy to use Redis by mapping Redis data structures straight to your code, allowing for a more natural and intuitive way of working with Redis from Node.js applications. It introduces new data structures such as JSON and Search and Query, which enable the creation of document databases that combine hierarchy and search capabilities. The library provides a fluent interface for creating, reading, updating, and deleting entities, as well as searching for them. With this library, developers can easily add Redis to their Node.js applications without needing to know complex Redis commands.

The cloud has been around for over 10 years, transforming our lives with its exponential innovation. Despite this, organizations still struggle with cloud operations and migration, with 1 in 3 migrations failing to deliver any tangible value. This is partly due to the lack of a clear definition of when an organization is "in the cloud," as well as the complexity of data management across multiple clouds. However, companies that successfully migrate and adopt cloud technology are evolving to abstract infrastructure burdens and leveraging managed services to optimize their use of cloud resources. Ultimately, a new approach to the data layer is needed to simplify and accelerate digital transformation.

This Qwiklabs module, in collaboration with Google Cloud, teaches users how to run Redis Enterprise and a Knative serverless application on an Anthos Bare Metal cluster deployment, showcasing the capabilities of Redis anywhere and on any cloud. Redis is an open-source in-memory data structure store designed for fast response times, while RediSearch enables retrieving hashes based on their contents. The module guides users through installing the Redis Enterprise Operator and creating a Knative serverless application, providing a cloud-native means of managing Redis Enterprise instances in a Kubernetes environment. This lab offers hands-on knowledge about Google Cloud and Redis Enterprise, empowering users to build real-time apps and scale out based on traffic from performance testing workloads.

With Redis simplifies your architecture by providing a unified real-time data platform that reduces complexities and costs associated with managing multiple specialized services like Kinesis, DynamoDB, ElastiCache, TimeStream, and OpenSearch. A single system using native features like streams, pub/sub, Lists, TimeSeries, Gears, and RedisJSON can replace the need for numerous experts to manage individual systems, resulting in a blazing fast and cost-effective solution.

As your IT infrastructure grows, managing logs can become complex and tedious, especially with the rise of cloud-based applications. To simplify this process, Alexis Gardin created Logub, an application that uses Redis to collect, explore, and analyze application logs on-premises or open-source. Logub is built using RediSearch's ability to efficiently explore and analyze logs across different locations. The application allows users to interact with a DEMO app, generate logs, request them, and carry out comprehensive searches using RediSearch's QueryBuilder. Logs can be stored using the Fluentd Redis Plugin, which stores each log in a HashSet type of Redis. To track indexed fields, a 'schema' object is used, allowing users to add or remove fields as needed. Logub provides an innovative solution for managing logs, making it easier to search, gather, and store logs with ease.

AWS has made Redis Enterprise Cloud available in its marketplace, allowing customers to procure it with a simplified consolidated bill and flexible on-demand pricing that counts toward their EDP consumption. This new procurement method provides joint customers with a modernized purchasing experience and empowers them to provision according to changing business needs without budget commitment. The service offers unified billing, industry-leading functionality, and fully managed Redis services in the cloud, freeing customers from operational complexities and enabling them to quickly and cost-effectively procure Redis Enterprise Cloud.