Redbus Journey: DragonflyDB Evaluation & Adoption (2025–2026)

1. Summary

The transition from AWS Elasticache(Redis) to DragonflyDB at Redbus was initiated to solve hardware underutilization and improve performance-per-dollar. As of early 2026, the migration is nearing completion with significant gains in memory efficiency, throughput and dollar savings.

2. Dragonfly Architecture

Dragonfly is a Redis- and Memcached-compatible, high-performance in-memory data store designed as a drop-in replacement for Redis, built from the ground up with a modern architecture.

• Thread-per-core, shared-nothing architecture: Leverages all CPU cores efficiently through vertical scaling. The dataset is sharded, with each shard managed by a dedicated thread, minimizing locks and synchronization.
• Efficient inter-thread communication: Achieved through message passing, avoiding shared memory contention.
• Custom fiber-based framework: Built on C++ stackful fibers (similar to Go’s goroutines) to abstract I/O operations and manage concurrency within each thread.
• Snapshotting with io_uring: Uses Linux’s efficient async I/O interface instead of Redis traditional fork + copy-on-write (COW), significantly improving I/O performance.
• Data structure enhancements:
• DashTables: Replaces traditional Redis hashtables for better memory efficiency and performance.
• B+ Trees: Used for sorted sets to optimize indexing and range queries.
• SSD-based tiering: Introduced specifically for string values to allow datasets to exceed available RAM.
• Command support: While Dragonfly supports most Redis commands, not all are implemented yet.
• Licensing: Released under Business Source License 1.1 (BSL 1.1) — free for internal use, but restricted from being offered as a managed service.

3. Setup & Performance Benchmarking: Dragonfly vs. Valkey vs. Redis

Although Redis and Valkey are largely identical, we enabled I/O threading in Valkey (with 6 threads) to evaluate how Redis/Valkey compares to Dragonfly, which was run with 8 threads (utilizing all available CPU cores).

Infrastructure

• Cache Server: 8-core, 64 GB RAM
• Client machine: 4-core, 8 GB RAM
• Tool: memtier_benchmark (Redis Labs)

Techstack Versions

• Redis: Community Edition May 2025
• Valkey: Latest version May 2025
• Dragonfly: Community Edition 1.29.x April 2025

3.1 Benchmark : String SET & GET

Conclusion: Dragonfly outperforms Redis and Valkey (except in p99.9 latency).

Dragonfly vs Redis

• Winner: Dragonfly
• SET: ~239% higher operations/sec, ~70% lower average latency, ~29% less RAM than Redis
• GET: ~200% higher operations/sec, ~67% lower average latency, ~29% less RAM than Redis
• Snapshot: ~62% smaller snapshot size than Redis
• CPU: higher CPU (538%–843%) than Redis.

Valkey vs Redis

• Winner: Valkey (except for snapshot size)
• SET: ~107% higher operations/sec, ~51% lower average latency, slightly higher RAM than Redis
• GET: ~109% higher operations/sec, ~52% lower average latency, ~1% less RAM than Redis
• Snapshot: ~60% larger snapshot size than Redis
• CPU: higher CPU (506%–681%) than Redis.

Dragonfly vs Valkey

• Winner: Dragonfly (except for p99.9 latency)
• SET: ~64% higher operations/sec, ~38% lower average latency, ~29% less RAM than Valkey
• GET: ~44% higher operations/sec, ~30% lower average latency, ~29% less RAM than Valkey
• Snapshot: ~76% smaller snapshot size than Valkey
• CPU: higher CPU (5%–21%) than Valkey.
• Valkey outperforms Dragonfly for SET by 7% in p99.9 latency, for GET by 16% in p99.9 latency (reasons unknown).

3.2 Benchmark : Sorted Sets (ZADD, ZRANGEBYSCORE)

Conclusion: Dragonfly outperforms Redis (except in ZADD p99.9 latency) and Valkey (except in ZADD p99 & p99.9 latency).

Dragonfly vs Redis

• Winner: Dragonfly (except for ZADD p99.9 latency)
• ZADD: ~220% higher operations/sec, ~67% lower average latency, ~43% less RAM than Redis
• ZRANGEBYSCORE: ~183% higher operations/sec, ~64% lower average latency, ~40% less RAM than Redis
• Snapshot: Slightly higher snapshot size than Redis
• CPU: higher CPU (189%–438%) than Redis.
• Redis outperforms Dragonfly for ZADD by 341% in p99.9 latency (reasons unknown).

Valkey vs Redis

• Winner: Valkey (except in snapshot size)
• ZADD: ~99% higher operations/sec, ~49% lower average latency, ~14% more RAM than Redis
• ZRANGEBYSCORE: ~118% higher operations/sec, ~54% lower average latency, ~10% more RAM than Redis
• Snapshot: Slightly higher snapshot size than Redis
• CPU: higher CPU (241%–472%) than Redis.

Dragonfly vs Valkey

• Winner: Dragonfly (except in p99 & p99.9 latency)
• ZADD: ~61% higher operations/sec, ~35% lower average latency, ~50% less RAM than Valkey
• ZRANGEBYSCORE: ~30% higher operations/sec, ~22% lower average latency, ~45% less RAM than Valkey
• Snapshot: Slightly higher snapshot size than Valkey
• CPU: lower CPU (6%–15%) than Valkey.
• Valkey outperforms Dragonfly for ZADD by 97% in p99 latency and ~645% in p99.9 latency (reasons unknown).

3.3 Benchmark : Hash (HSET, HMGET)

Conclusion: Dragonfly outperforms Redis and Valkey (except for HMGET in p99 & p99.9 latency).

Dragonfly vs Redis

• Winner: Dragonfly
• HSET: ~258% higher operations/sec, ~72% lower average latency, ~5% less RAM than Redis
• HMGET: ~102% higher operations/sec, ~50% lower average latency, ~5% less RAM than Redis
• Snapshot: ~84% lower snapshot size and ~94% lower snapshot time than Redis
• CPU: higher CPU (250%–605%) than Redis.

Valkey vs Redis

• Winner: Valkey (except in snapshot time)
• HSET: ~83% higher operations/sec, ~45% lower average latency, ~3% more RAM than Redis
• HMGET: ~92% higher operations/sec, ~47% lower average latency, ~3% more RAM than Redis
• Snapshot: same snapshot size but ~1% higher snapshot time than Redis.
• CPU: higher CPU (241%–516%) than Redis.

Dragonfly vs Valkey

• Winner: Dragonfly (except in p99 & p99.9 latency)
• HSET: ~95.3% higher operations/sec, ~49% lower average latency, ~8% less RAM than Valkey
• HMGET: ~5% higher operations/sec, ~4% lower average latency, ~8% less RAM than Valkey
• Snapshot: ~83% lower snapshot size and ~94% lesser snapshot time than Valkey
• CPU: higher CPU (3%-14%) than Valkey.
• Valkey outperforms Dragonfly for HMGET by 16% on p99 latency and 32% for p99.9 latency (reasons unknown).

3.4 Benchmark : LIST (LPUSH, LRANGE)

Conclusion: Dragonfly outperforms Redis (except for snapshot size & time) and Valkey (except in p99 & p99.9 latency, snapshot size & time).

Note: The benchmark involved pushing 8 crore events into a list — an unrealistic scenario, solely done for testing purposes.

Dragonfly vs Redis

• Winner: Dragonfly (except for snapshot size and time)
• LPUSH: ~194% higher operations/sec, ~65% lower average latency, ~1% less RAM than Redis
• LRANGE: ~38% higher operations/sec, ~26% lower average latency, ~11% less RAM than Redis
• Snapshot: ~1865% larger snapshot size and ~2914% higher snapshot time than Redis
• CPU: higher CPU (237%–321%) than Redis.

Valkey vs Redis

• Winner: Valkey (except in snapshot time)
• LPUSH: ~125% higher operations/sec, ~55% lower average latency, slightly higher RAM as Redis
• LRANGE: ~79% higher operations/sec, ~44% lower average latency, ~15% less RAM than Redis
• Snapshot: slightly higher snapshot size and time compared to Redis.
• CPU: higher CPU (309%–480%) than Redis.

Dragonfly vs Valkey

• Winner: Dragonfly (except in snapshot and LRANGE latency)
• LPUSH: ~31% higher operations/sec, ~21% lower average latency, ~1% less RAM than Valkey
• LRANGE: ~23% lesser operations/sec, ~32% higher average latency, ~5% more RAM than Valkey
• Snapshot: ~1862% larger snapshot size and ~2937% higher snapshot time than Valkey
• CPU: higher CPU (17%-27%) than Valkey.
• Valkey outperforms Dragonfly for LPUSH its 18% in p99 latency, for LRANGE its 618% on p99 latency and 415% for p99.9 latency (reasons unknown).

3.5 Benchmark : Set (SADD, SISMEMBER)

Conclusion: Dragonfly outperforms Redis (except for snapshot size & time) and Valkey (except for SADD in p99 latency, snapshot size & time).

Note: The benchmark involved pushing 4 million events into a set — an unrealistic scenario, solely done for testing purposes.

Dragonfly vs Redis

• Winner: Dragonfly (except in snapshot time)
• SADD: ~180% higher operations/sec, ~63% lower average latency, ~13% less RAM than Redis
• SISMEMBER: ~182% higher operations/sec, ~63% lower average latency, ~3% less RAM than Redis
• Snapshot: Slightly higher snapshot size and 42% to 46% higher snapshot time than Redis
• CPU: higher CPU (194%–479%) than Redis.

Valkey vs Redis

• Winner: Valkey (except in snapshot time)
• SADD: ~122% higher operations/sec, ~55% lower average latency, slightly lower RAM than Redis
• SISMEMBER: ~149% higher operations/sec, ~59% lower average latency, ~5% higher RAM than Redis
• Snapshot: similar snapshot size but 3% higher snapshot time than Redis.
• CPU: higher CPU (226%–484%) than Redis.

Dragonfly vs Valkey

• Winner: Dragonfly (except in SADD p99 latency)
• SADD: ~26% higher operations/sec, ~18% lower average latency, ~12% less RAM than Valkey
• SISMEMBER: ~13% higher operations/sec, ~10% lower average latency, ~7% less RAM than Valkey
• Snapshot: Slightly higher snapshot size and 47% to 51% higher snapshot time than Valkey
• CPU: lower CPU (1%-10%) than Valkey.
• Valkey outperforms Dragonfly for SADD by 17% in p99 latency (reasons unknown).

3.6 Final summary

Disclaimer: We are publishing our findings as found in the setup and the tech stack versions listed above.

Performance

• Across benchmarks, Dragonfly overall outperforms Redis and Valkey.
• Dragonfly has consistant p99 latencies than Valkey or Redis.
• Snapshot sizes are sometimes larger, likely due to extreme benchmark scenarios involving heavy List and Set usage but nonetheless needs to be checked with the dragonfly team.

Scalability

• Dragonfly scales impressively with more CPU cores, challenging Redis single-threaded, horizontally-scaled philosophy.
• Dragonfly supports horizontal scaling as well, enabling performance improvements through both vertical and horizontal scaling.

Setup & Support

• Setting up Dragonfly is straightforward — it’s a lightweight binary with a configuration process similar to Redis. You can use redis-cli to interact with it.
• While Dragonfly doesn’t yet support the full set of Redis commands, its feature set is actively growing and already covers the majority of use cases within Redbus.

Stability & Reliability

• Dragonfly is production-ready and already adopted by many companies, however it remains a relatively new and evolving player in the in-memory database space — unlike Redis, which is a mature, battle-tested solution with years of stability and community trust.

Conclusion

Overall Dragonfly is a strong modern alternative to Redis that remains bottlenecked by its single-threaded nature., offering excellent hardware utilization and performance-per-cost. That said, we recommend evaluating it thoroughly against your specific workload before adoption.

4. Migration & Adoption Status (March 2026)

Redbus decided to migrate to Dragonfly cloud. Redbus has seen aggressive adoption across its global cloud infrastructure with ~90% of total caches migrated.

5. Cost Savings & Optimization Opportunities

By migrating to Dragonfly cache we have targeted to save atleast ~40% compared to our earlier spends on AWS Elasticache(with a savings plan). Savings are much higher compared to on-demand costs.

6. Contributors

• Harsha Nanjundaiaha
• Nitish P
• Swatik J Shetty
• Tanveer M

7. References

1. https://www.dragonflydb.io/blog/redis-and-dragonfly-architecture-comparison
2. https://github.com/dragonflydb/dragonfly
3. https://redis.io/
4. https://github.com/valkey-io/valkey