MongoDB vs Redis

Redis Market Stance

Redis occupies a singular position in the modern data infrastructure stack. It is simultaneously the most popular in-memory data store in the world, a foundational component of virtually every high-traffic web application built in the last decade, and — through Redis Ltd. — a commercial enterprise software company executing a deliberate transition from open-source community project to cloud-delivered subscription business. Understanding Redis requires holding both of these identities in mind at once, because the tension and synergy between them defines the company's competitive dynamics, commercial strategy, and long-term trajectory. The Redis project was created in 2009 by Italian developer Salvatore Sanfilippo, who was building a real-time web analytics tool and needed a data structure server that could handle high-velocity reads and writes with sub-millisecond latency — something no existing database could deliver at the time. His solution was to build an in-memory key-value store that kept all data in RAM rather than on disk, sacrificing persistence in exchange for extraordinary speed. The result — Redis, which stands for Remote Dictionary Server — was open-sourced almost immediately and achieved rapid community adoption, becoming the default answer to the question of how to add caching, session management, or real-time data handling to a web application. The speed of Redis adoption through the 2010s reflects a genuine architectural insight: as web applications scaled to millions of concurrent users, the bottleneck was not compute but data access latency. Traditional relational databases optimized for durability and consistency but could not serve sub-millisecond reads at scale. Redis solved this problem elegantly, and developers — who could install and configure it in minutes — adopted it enthusiastically. By the mid-2010s, Redis had become one of the most popular databases in the world by usage volume, consistently ranking in the top five of DB-Engines' database popularity ranking alongside Oracle, MySQL, PostgreSQL, and SQL Server — a remarkable achievement for a project built by a single developer. Redis Labs (later renamed Redis Ltd.) was founded in 2011 to build a commercial business around the open-source Redis project. The company's initial strategy was straightforward: offer managed Redis services — Redis as a Service — that removed the operational burden of running Redis in production at scale. Managing Redis at scale is non-trivial: it requires expertise in memory management, replication configuration, cluster topology, persistence tuning, and high-availability architecture that most engineering teams prefer not to build in-house. Redis Labs charged for this operational expertise and infrastructure, creating a viable commercial model on top of a freely available open-source product. The commercial evolution of Redis Ltd. has passed through several strategic phases. The early managed service phase established the company's cloud credentials and built an enterprise customer base. The Enterprise product phase — introducing Redis Enterprise, an on-premise and cloud distribution of Redis with additional capabilities including active-active geo-distribution, advanced security, and automated failover — expanded the addressable market from cloud-native companies to regulated industries and large enterprises with on-premise infrastructure requirements. The current cloud-first phase is executing a deliberate migration of the customer base toward Redis Cloud, a fully managed, multi-cloud Redis service that generates the recurring subscription revenue profile most valued by enterprise software investors. A pivotal — and controversial — strategic decision came in 2024 when Redis Ltd. changed the license of the core Redis software from the BSD open-source license to a dual license under the Redis Source Available License (RSALv2) and the Server Side Public License (SSPLv1). This change prevented cloud providers — specifically Amazon Web Services, Microsoft Azure, and Google Cloud — from offering Redis as a managed service without entering a commercial agreement with Redis Ltd. The licensing change was a direct response to what Redis Ltd. characterized as value extraction by hyperscalers: AWS, Azure, and Google had each built their own managed Redis-compatible services (ElastiCache, Azure Cache for Redis, and Memorystore respectively), capturing the high-value managed service opportunity that Redis Ltd. had pioneered without contributing to its commercial sustainability. The licensing change generated significant controversy in the open-source community and prompted a fork: the Linux Foundation, backed by AWS, Google, and a coalition of Redis contributors, launched Valkey — a BSD-licensed fork of Redis — within weeks of the license announcement. This fork represents the most significant competitive and community challenge in Redis's history, and its implications for Redis Ltd.'s long-term commercial position are still unfolding. The Redis story, as of 2025, is inseparable from the Valkey fork and the broader question of whether open-source commercialization models can survive in an era of hyperscaler platform power. Beyond the licensing controversy, Redis has expanded its technical scope significantly. Originally a simple key-value store, Redis now supports a rich set of data structures — strings, lists, sets, sorted sets, hashes, streams, geospatial indexes, and probabilistic data structures like HyperLogLog and Bloom filters — that enable use cases far beyond basic caching. Redis Streams, introduced in version 5.0, added a persistent, consumer-group-aware message queue capability that competes directly with Apache Kafka for event-driven architecture use cases. Redis Search and Redis JSON, available as modules in Redis Enterprise and Redis Cloud, extend the platform into full-text search and document store territory respectively. The emergence of AI and machine learning as dominant enterprise computing paradigms has opened a new and significant chapter for Redis. Vector similarity search — the ability to store and query high-dimensional vector embeddings generated by machine learning models — has become one of the fastest-growing database use cases as enterprises build retrieval-augmented generation (RAG) applications on top of large language models. Redis Vector Library and the vector search capabilities embedded in Redis Cloud and Redis Enterprise position Redis as a natural infrastructure choice for AI application developers who are already using Redis for caching and need a low-latency vector store for semantic search and recommendation systems. This AI adjacency is the most important new commercial opportunity in Redis's current strategy.