Dragonfly 2025 in Review: The Year of Growth, Consolidation, and Maturation

As 2025 comes to a close, we reflect on a pivotal year for Dragonfly, defined by decisive product evolution and strategic cloud service enhancement. This period has been marked by significant technical releases and platform developments that have solidified Dragonfly’s position as the next-generation in-memory data store and a robust, production-ready solution for modern, data-intensive applications. The year’s focus was squarely on advancing core technology, expanding cloud capabilities, and delivering measurable value at scale.

This momentum is reflected by the Dragonfly project surpassing 29,500 GitHub stars, a key indicator of its growing adoption and recognition as a critical piece of data infrastructure.

From major feature releases in the community edition to the expansion and refinement of Dragonfly Cloud, this blog post recaps the key milestones of 2025 at Dragonfly. We will examine how these advancements provide developers, teams, and organizations with a powerful, real-time data layer to modernize existing applications and build new, high-performance ones. Join us as we review a year of concrete technical progress and explore the foundation it creates for the future.

Engineering Excellence: The Community Edition in 2025

Dragonfly is committed to providing a powerful, self-hostable Community Edition, which is a source-available engine that is free to use forever. In 2025, our engineering focus was on hardening this core, transforming advanced features from experimental into robust, production-ready ones. This year’s releases were characterized by deep refinements that delivered tangible gains in capability, performance, and operational stability.

A Leap Forward in Search Capability

The native Dragonfly Search functionalities received significant attention, moving beyond foundational support to deliver a refined query experience. Key improvements included:

• Performance Optimizations: Enhanced indexing and query execution mechanisms for lower latency and higher throughput on complex searches.
• Expanded Query Flexibility: The introduction of infix and suffix search support for tags enables more flexible and intuitive data lookup patterns critical for application features like user mentions or hashtag exploration.
• Enhanced Geo-Spatial Indexing: The addition of GEO index solidifies Dragonfly as a top choice for location-aware applications.

Read More: Dragonfly’s Inverted GEO Index

The Maturation of SSD Tiering

Another achievement for the year was the maturation of our SSD data tiering feature. Once an experimental feature, it is now considered stable and ready for production use within the Community Edition. This evolution provides a critical cost-performance benefit:

• Transparent Data Management: Enables large datasets beyond the limit of RAM by automatically offloading less frequently accessed data to SSDs.
• Reduced TCO: Dramatically lowers the cost per gigabyte for large datasets without sacrificing the sub-millisecond latency for active data, making Dragonfly viable for new classes of data-intensive workloads.

Read More: Announcing Dragonfly SSD Data Tiering

Other Refinements and Expansion

Beyond these notable features, 2025 was a year of widespread polish and extension:

• Performance & Stability: Numerous under-the-hood optimizations reduced memory overhead and improved throughput across a wide range of command patterns.
• Command Coverage: Continued expansion of compatibility with the evolving Redis ecosystem, with support for additional commands, ensuring smoother migrations.
• Bug Fixes & Reliability: A consistent focus on squashing bugs and improving overall system resilience for demanding, long-running deployments.

Together, these advancements ensure that the Dragonfly Community Edition is not just a Redis/Valkey alternative but a compelling next-generation, enterprise-grade foundation for any team looking for an in-memory data store that can handle the heaviest workload in the wide-context AI era.

The Cloud Frontier: Availability, Power, and Control

2025 was the year Dragonfly Cloud transformed from a powerful managed service into a comprehensive, enterprise-ready platform available anywhere you run your applications. This expansion was marked by three major pillars: universal availability, unmatched and autonomous scalability, and a new tier designed for the most demanding workloads.

Ubiquitous Platform Availability

Dragonfly Cloud achieved a major milestone in 2025 with availability across all three major cloud platforms: AWS, Google Cloud, and Microsoft Azure. This multi-cloud strategy ensures teams can deploy a high-performance, Dragonfly-powered data layer in their environment of choice.

To further streamline procurement and operations, Dragonfly Cloud became purchasable directly via the AWS and Google Cloud Marketplace. This integration simplifies management and billing, allowing you to manage your Dragonfly Cloud resources under unified billing.

Dragonfly Swarm: In-Memory Data without Limits

A landmark release for the year was Dragonfly Swarm, a revolutionary clustering solution that redefines horizontal scaling for in-memory data. Announced in February and now a core feature of Dragonfly Cloud, Swarm addresses the fundamental limitations of legacy solutions.

Read More: Dragonfly Swarm

Unlike traditional in-memory data stores that must scale out horizontally at small sizes, Dragonfly’s multi-threaded architecture allows a single shard to scale vertically first, fully utilizing powerful cloud instances. Only when limits are reached does Swarm enable seamless horizontal expansion. This “vertical + horizontal” approach minimizes cluster complexity, reduces overhead, and provides a frictionless path to scaling beyond 100TB of memory and 100 million RPS, delivering both radical simplicity and unmatched cost efficiency for terabyte-scale workloads.

Dragonfly Cloud Enterprise: For the Heaviest Workloads

In August, we introduced Dragonfly Cloud Enterprise, a new tier engineered for organizations with the most stringent requirements for scale, security, and reliability. This tier is built around a bold promise: a guaranteed minimum of 30% cost savings compared to your existing managed in-memory services.

Read More: Dragonfly Cloud Enterprise

Beyond cost savings, Dragonfly Cloud Enterprise also packages powerful capabilities critical for large-scale operations:

• Bring-Your-Own-Cloud (BYOC): Run the fully managed Dragonfly service within your cloud account (AWS, GCP, or Azure) for enhanced data control and compliance.
• Intelligent Autoscaling: Dynamically adjusts resources to match traffic patterns, ensuring performance during spikes and cost-efficiency during lulls.
• Robust Data Resilience: Features include multi-region backups for disaster recovery and intelligent read replicas to optimize read-heavy workloads.
• High-Touch Enterprise Support: Provides 24/7 assistance directly from the Dragonfly engineering team.

A Seamless Management Experience

Supporting these powerful new capabilities is a brand-new Dragonfly Cloud console. This redesigned interface provides a more intuitive and seamless experience for managing every aspect of your data stores: from provisioning Swarm clusters and configuring VPC peering to managing team member access and backups, all within a single, cohesive dashboard.

Trusted by Industry Leaders

The performance and versatility of Dragonfly Cloud attracted a diverse range of customers scaling their mission-critical applications in 2025. We were proud to welcome organizations like Instacart, the popular online grocery delivery service, and ShareChat, India’s leading social media platform. Their adoption demonstrates Dragonfly Cloud’s ability to meet the extreme performance, reliability, and scaling requirements of both hyper-growth social networks and high-volume AI/ML platforms.

Together, these 2025 releases solidify Dragonfly Cloud as a unified, high-performance data platform capable of meeting the diverse needs of startups and global enterprises alike, wherever they choose to build.

Community in Motion: A Global Connection

Our presence and engagement within the global technology community reached new heights in 2025. Through a mix of flagship events, major industry conferences, and grassroots meetups, we connected with thousands of developers, architects, and engineers to showcase Dragonfly’s role in the modern data stack.

Hosting the Modern Data Infrastructure Summit

This year, we proudly hosted the first-ever Modern Data Infrastructure Summit. This event brought together the brightest minds and leading innovators across the data ecosystem. We were honored to partner with pioneering companies like MotherDuck, ScyllaDB, GreptimeDB, Redpanda, and TiDB for insightful discussions and deep technical sessions. The summit served as a powerful forum to explore the future of real-time data, high-performance databases, and cloud-native architectures, solidifying the collaborative spirit that drives our industry forward.

Spotlight on Major Industry Conferences

The Dragonfly team took our cloud-native and high-performance message directly to the heart of the industry on two major stages:

• KubeCon North America 2025: We emphasized Dragonfly’s native integration with Kubernetes, demonstrating its fit as the stateful, high-performance data layer for any cloud-native applications.
• AWS re:Invent 2025: As a key launch platform for our expanded cloud availability, our presence at re:Invent highlighted Dragonfly Cloud’s deep integration with the AWS ecosystem and its enterprise-ready capabilities for the most demanding in-memory workloads.

Widespread Community Engagement

2025 was also filled with active participation across a wide spectrum of gatherings. We deeply appreciate the collaboration and speaking opportunities provided at numerous events throughout the year, including: Current Bengaluru, AI Council, AWS Summit Bengaluru, Boston Kubernetes Meetup, Toronto Open Source Meetup, TorontoJS, Nebraska.Code, KubeCon India, AWS Summit Toronto, TiDB SCaiLE, and P99 CONF.

From large-scale summits to friendly local meetups, both online and in-person, each event was a valuable opportunity to learn, share, and connect with the vibrant communities that are building the future of data infrastructure and AI-powered applications.

Looking Ahead to 2026

The milestones of 2025, including the limitless scale of Dragonfly Swarm, the enterprise-grade power of Dragonfly Cloud, and our ever-growing global community, have set a strong foundation. As we look to the new year, our focus sharpens on deepening this impact. We will continue to push the boundaries of performance and scalability while making the Dragonfly experience more seamless and accessible for every team, from ambitious startups to the largest global organizations.

We are extremely grateful for the trust of our users, the contributions of our open-source community, and the partnerships forged this year. Thank you for making 2025 a transformative chapter in the Dragonfly story.

In this modern AI era, the journey is only accelerating. Stay tuned for our updates in 2026. To be a part of it, join the conversation on our Discord server, contribute to Dragonfly on GitHub, and connect with us at our next event. We can’t wait to build the future of in-memory data with you.