MongoDB takes aim at serverless cloud and offers graph functionality with latest release

Update also includes a host of security enhancements

MongoDB 3.4, released today, makes available new functionality such as graph capabilities in the NoSQL database.

MongoDB is keen to push the word "multimodel" to describe the additional functionality in its NoSQL database. Multimodel is a fancy world for a general-purpose database that provides integrated back-end support for many different data models.

Version 3.4 offers new features, plus refinements to existing ones. There are few major surprises here as the company chose to preview them at the MongoDB World event in June.

One of the new features is additional functionality allowing users to edit documents from the Compass GUI. The latest version also includes an "all-new SQL interface", according to MongoDB. The idea is to allow MongoDB to be used as a data source for BI systems without complex ETL operations.

In addition, to ensure it remains relevant at the cutting edge of analytics, MongoDB has upgraded its Spark connector to work with Apache Spark 2.0, the latest version of the popular streaming analytics platform.

Then there are Zones, the fully elastic database partitioning capability built into MongoDB's admin tools designed for multi-region deployments. Zones are designed to allow flexibility to database administrators to as to whether data is hosted in local data centres to meet data sovereignty criteria or on tiered hardware for performance.

As anticipated, version 3.4 also offers graph capabilities and faceted navigation. Graph databases are useful for analysing things like social networks and friend-of-a-friend relationships. They are also helpful for modelling and optimising of complex physical networks. Faceted navigation will be familiar to users of Ebay as the boxes in the left-hand navigation pane that allow you to refine your search criteria.

There are security improvements too, as MongoDB offers read-only views to reduce the risk of data being copied or manipulated by the wrong people, and pattern matching to alert operators to the possible presence of sensitive data that's in a place where it shouldn't be.

Meanwhile MongoDB's database-as-a-service (DBaaS) offering, Atlas, takes aim at the latest cloud innovation known as Function as a Service (FaaS), or serverless computing.

"Support for AWS VPC peering is being added to MongoDB Atlas, enabling users to create an extended, private network that connects their application servers, and services such as AWS Elastic Beanstalk and AWS Lambda to their MongoDB Atlas databases without using public IP addresses that could compromise security," the company says in a statement.

Also in the cloud, the company's Ops Manager tool introduces native Cloud Foundry integration, making it easier to provision and manage databases on that platform.

Non-English language users will be hoping that the latest version delivers on MongoDB's promises to deliver collated sorting, i.e. sorting on accented characters, a feature first requested back in 2010.