At the end of April there was a new release of ONgDB that begins a fully open era for graphs. We had the opportunity to celebrate this release at The Honest Company last night with the Graph Database LA Meetup group where I shared many of the highlights from the release. The latest release ushers in a new era of scalable yet reliable graph database technology with, this version of ONgDB based on a completely redesigned architecture that offers enhanced developer productivity, and varying deployment options at a massive scale.
Here’s what to be expected with the new ONgDB:
- Redesigned internals that eradicates limits on node numbers and restoration of indexed and stored properties and relationships.
- Official support for language drivers via Bolt binary protocol and Java Stored Procedures support, while enabling full-stack developers for powerful application creation.
- Streamlined deployment structure and configuration for deploying ONgDB in the cloud or on premise.
Here’s an in-depth look of what’s new in the latest version:
- Unlimited Graph Storage
By far the biggest headline in the release. Graph to size infinite – challenge accepted! Dynamic pointer compressions expands the available address of ONgDB as needed, making it possible to house graphs regardless of size. Such features make ONgDB even more enterprise-ready, by complementing its existing scale-out clustering features.
- Enhanced Cost-based Optimizer
This is a huge one for us because most of the Geequel queries we write are complex MERGE operations so we need as much write performance as possible. A cost-based optimizer has been enhanced by adding support for write queries. The new parallel indexes capability within the optimizer also allows for swifter performance population of indexes.
- Language Drivers & Bolt
Bolt is great for ONgDB developers because it means better performance of the applications they build all the way around and enables them to go bigger and do more with ONgDB. Bolt is a connection-driven protocol for graph access. It utilizes a portable binary encoding over web sockets or TCP for lower latency and enhanced throughput. It comes with built-in security that enhances both graph database performance and developer experience.
- Java Stored Procedures
These new and powerful performance facility offers low-level and direct graph access, giving you a way to conduct an imperative code when you want to conduct complex work within the database. ONgDB comes bundled with built-in procedures as part of the APOC project. There are some very useful procedures in the APOC project so you should definitely check it out. One that stood out to me as immediately useful is the last one in the list that makes periodic commit available for use outside of LOAD CSV.
- Cloud Deployable
This restructure was a pretty major change and getting all the updates for GraphGrid to support the latest ONgDB was a bit of an undertaking but we got it all in place and do like the new organization. The new file, config and log structures in the latest ONgDB are made for streamlining operations. A notable change is to transfer from multi-config sites to one of name-spaced files. A config migration utility is supplied for seamless transition.
The result will let you easily operate ONgDB on premise and on deployment instances at a massive scale while fostering ONgDB deployment just about anywhere.
ONgDB is the next big evolution in graph database technology and now introduces a fully open era for graphs. ONgDB continues to define and lead the fully open source graph database market and we’re excited to be making massive scale graph solutions possible.