After 6 months of development we are happy and excited to announce the fully production ready ArangoDB 3.0 today! Get ArangoDB 3.0 now!

We designed ArangoDB as a native multi-model DB from the first line of code. By providing three major NoSQL data models in one technology the ArangoDB team wants to fulfill its mission to simplify data work. With ArangoDB 3 we believe that our users will come more than one step closer to a dramatically simpler way to create their applications.

The current constraint isn’t hardware anymore. Computing power is spinned with a few clicks. The current constraint is software. Software is written and operated by people. It’s now all about making people as informed, enabled and efficient as possible. The team behind ArangoDB wants to contribute to this new era. By providing a complete overhauled cluster, streamlined graph functionalities, persistent indexes, a new binary storage format and simplified extensibility via our Foxx framework. With these upgrades we created a solid ground to scale for our community.

We want to thank all of you who tested the alpha, beta and release candidates! Your feedback and contributions made 3.0 the best ArangoDB ever! Thank you so much!

 

Main improvements in ArangoDB 3.0

ArangoDB 3.0 comes with largely improved cluster capabilities. You can now build self-organizing clusters which are managed by ArangoDB’s new agency. Multi-Master setups, synchronous replication and the new shared-nothing architecture with auto-failover just made deployment and maintenance of clusters very easy. Read more about our cluster improvements .

All ArangoDB graph functions are now natively and deeply integrated into AQL which makes the usage of our graph capabilities way easier. All graph features are now much more flexible and you can use any AQL based filter. We didn’t just achieve better performance with it but streamlined the whole process of working with graphs.

Persistent indexes via RocksDB is the first step of ArangoDB to persist indexes in general. With the new persistent index you get the same functionality as our skiplist-index provides thus, it is perfectly suitable for sorting and range-queries. During our 3.x development phase we will provide many other types of persistent indexes.

ArangoDB’s new binary storage format VelocyPack has been implemented. With VelocyPack ArangoDB is now able to store your data in a very compact way, reduce the overall memory-usage and directly access subvalues without decoding the whole document. With many other improvements we managed to increase ArangoDBs performance up to 5x while reducing memory usage by 20%. A detailed performance benchmark for 3.0 will follow soon!

If you liked Foxx before you will definitely fall in love with Foxx 3.0 – our JS framework for data-centric microservices. You can now use ArangoDB collections directly. No more need for repositories. Controllers have been replaced with nestable routers. We cleaned up Foxx and added a dedicated and  detailed documentation for Foxx for it. If you already use Express for Node.js you will directly feel at home with the redesigned express-like middleware.

Read a full list of the all upgrades in our alpha release note.

ArangoDB is used by teams around the globe. Matured companies like Demonware, Liaison or ICmanage leverage the power of our multi-model database and striving startups like FlightStats, Fedger.io or AboutYou love the freedom to adapt quickly to customer needs and prototype their ideas real fast. We are so happy to see a widespread adoption of ArangoDB across different company sizes, markets and use cases.

Get started with ArangoDB 3.0 today and use one of our 10 minutes tutorials!

Tutorial for Java
Tutorial for PHP
Tutorial for NodeJS

Our core team would love to hear your feedback on ArangoDB 3.0

Have fun playing around with 3.0!