Dragonfly

Featured

Dragonfly Cloud MCP Server announcement card. The Claude, Cursor, and VS Code logos connect through the Dragonfly logo with an MCP badge to a live datastore card named orders-cache, showing a real-time metrics chart.

Introducing Model Context Protocol (MCP) for Dragonfly Cloud

November 17, 2026

Dragonfly Cloud now has a hosted, OAuth-ready MCP server. Connect Claude, Cursor, or VS Code and ask about your datastores in plain English.

All posts

Dragonfly for AI frameworks | Cover Image

Dragonfly for AI frameworks: search, memory, and a Redis-compatible path to production

Dragonfly is a Redis-compatible backend for AI search, memory, and runtime in LangChain, LlamaIndex, Agno, Semantic Kernel, and more.

Dailymotion Case Study | Cover Image

How Dailymotion unified its key-value stores on Dragonfly and saved $450K a year

Dailymotion moved three key-value engines onto Dragonfly in seven regions, cut nodes up to 75%, and halved recommendation cache latency on Dragonfly 2.0.

Announcing-Dragonfly-2-0

Announcing Dragonfly 2.0

Three years after 1.0, Dragonfly 2.0 delivers 54% more throughput, 35% lower average latency, and 30-40% less memory. Here is what changed.

Infrastructure for the Agent Era

Infrastructure for the Agent Era

AI agents will outnumber humans by orders of magnitude. Why legacy in-memory data stores can't scale for the agentic era.

SSD Data Tiering on Dragonfly Cloud

SSD Data Tiering is now on Dragonfly Cloud

Dragonfly’s SSD Data Tiering feature allows it to store data on SSD drives, extending the potential size of a workload up to 8X times more.

From Dict to DashTable

From dict to DashTable: How Dragonfly cuts memory overhead by 40%

Redis doubles its hash table in memory before moving a single key. Dragonfly's DashTable grows one segment at a time and cuts memory overhead by 40%.

Dragonfly Wings

Stay up to date on all things Dragonfly

Join our community for unparalleled support and insights

Join
blog card for dragonfly large collection replication

How Dragonfly Improves Large Collection Replication

Dragonfly's new tagged chunks replication format doubles throughput and cuts P99 latency up to 87%. See how fine-grained locking removes stalls.

cover image

How Dragonfly Cuts Celery & Sidekiq Queue Memory by 3-4x with Dictionary Compression

Dragonfly has gained a new type of compression for lists. This new feature enables efficient compression of small records which are traditionally very hard to compress.

cover image

How Instacart Achieved 50% Better Performance with 70% Fewer Nodes

How Instacart migrated its ad-serving feature store from managed Valkey to Dragonfly, cutting cluster size ~70% and average and P99 latency 50% in weeks.

Dragonfly vs Valkey 9.0 Cover

Dragonfly vs. Valkey 9.0 on AWS Graviton: An Honest Head-to-Head

Valkey 9.0 closed real ground on I/O. We reran the benchmarks on m7g and c7gn.metal to find out where, and where Dragonfly still wins.

cover image

SSD Data Tiering Is Generally Available

Dragonfly SSD Data Tiering is now GA. Scale beyond RAM limits on a single node — no resharding, no code changes. The only source-available Redis-compatible store with SSD tiering.

One Dragonfly Instance, Ten Workloads: How ACL Database Selectors Work

One Dragonfly Instance, Ten Workloads: How ACL Database Selectors Work

Learn how Dragonfly's ACL database selectors let you run multiple isolated workloads on a single instance — without spinning up separate Redis instances per app.

rebalancing

Memory-Aware Rebalancing Is Now Automatic in Dragonfly Cloud

Most cluster rebalancing tools spread hash slots evenly by count, not by memory — which does nothing for a shard overloaded with large values. Dragonfly Cloud's shard memory balancing tracks utilization at the individual slot level and automatically redistributes slot...

no sentinels required

No Sentinels Required: How Dragonfly Cloud Cuts Failover to One Second

Dragonfly Cloud cuts Redis failover from 30+ seconds to 1 second by embedding failure detection directly into the data plane. No Sentinel required.

Redbus & Dragonfly

Why Redbus Skipped Valkey and Bet on a New Cache Architecture

Redbus, one of the world's largest travel ticketing platforms, migrated a large distributed cache from ElastiCache to Dragonfly for lower costs and better efficiency.

large network

Why Multi-Terabyte Redis Deployments Are Due for a Rethink

At a certain scale, Redis starts costing you more than just money.

Building a Self-Fuzzing CI Pipeline for Dragonfly

Building a Self-Fuzzing CI Pipeline for Dragonfly

This post covers how we went from ad-hoc manual fuzzing to a fully automated CI pipeline where an LLM generates targeted attack vectors for every pull request.

Vector Search Just Got Faster: Dragonfly v1.37 Delivers Up to 7x Throughput Gains and 65x Lower Latency

Vector Search Just Got Faster

Dragonfly v1.37 Delivers Up to 7x Throughput Gains and 65x Lower Latency.

Switch & save up to 80%

Dragonfly is fully compatible with the Redis ecosystem and requires no code changes to implement. Instantly experience up to a 25X boost in performance and 80% reduction in cost