Dragonfly 1.21 released today as the newest version of this modern, drop-in replacement to Redis and Memcached. This performance-optimized in-memory data store has added a few interesting features with the new release.

Last year with Dragonfly 1.13 it added [1]experimental SSD-based data tiering as another means of helping to boost the performance. With today's Dragonfly 1.21 release the SSD data tiering is now treated as alpha level. The "--prefix some/path/basename" option can be used for leveraging the SSD data tiering with this otherwise in-memory data store.

Dragonfly 1.21 also adds initial support for multi-tenant deployments. With this new feature it's possible to create multiple, separate, and isolated namespaces within Dragonfly. Each user can be associated to a single namespace. This multi-tenant feature is currently experimental and some features like defrag and replication aren't yet supported.

Downloads and more details on the other changes to find with the Dragonfly 1.21 release can be found via [2]GitHub .



[1] https://www.phoronix.com/news/Dragonfly-0.13-Released

[2] https://github.com/dragonflydb/dragonfly/releases/tag/v1.21.0