Users of the MongoDB NoSQL data store can now store graph databases in it. Today the company announced the introduction of MongoDB 3.4, and with it came a host of new features, including the ability to host graph databases.
Kelly Stirman, vice president of strategy and product marketing at MongoDB, said this update focuses on making MongoDB the one database a developer needs, rather than having to chain together a half dozen servers to cover various database needs. To this end, the company has been adding multi-model capabilities for some time.
Graph databases are just the latest in this path to multi-model. “We are not a best-in-class graph database, but we are adding in capabilities that we think cover most of what people do with graph,” said Stirman. “We think the appeal of this is that people don’t want to manage 15 different database technologies: They want to manage a couple. If they can get what they need out of Mongo, then they don’t have to stand up Neo4j.”
(Related: A tour of Quip’s architecture by Bret Taylor)
This move into subsuming the duties of other data stores was mirrored by the introduction of in-database search functionality, which Stirman said should eliminate the need to dumb MongoDB contents into something like Elastic Search or Solr.
Another major change to version 3.4 is the introduction of collation. Previously, MongoDB only supported UTF-8 for data labels, meaning sorting information could yield errors related to where accented letters are placed alphabetically. Now, MongoDB supports more than 100 languages, meaning alphanumeric sorting can be performed as it would in a developer’s native language.
MongoDB management and clustering tools were also updated for version 3.4. Said Stirman, “We’ve continued to improve the leader election process for failovers to be faster and more reliable in this release. We’ve introduced a new feature for linearized reads. Now developers can offer guarantees around consistencies they could never make before.”
This dovetails with the introduction of Zones. “If I have copies of data on different servers in Europe, North America, the East Coast, the West Coast, I want to make sure users are accessing their copy of the data locally so it’s always fast and gives them a great experience,” said Stirman. “You want to keep the data near your users at all times, if possible. With the Zones, you can operationally say, ‘These users are close to these datacenters, make sure their reads and writes are always running close to them.’ ”
Compass, the MongoDB management tool, has also received updates, which include a refined UI to handle Zones. MongoDB can also be administered from third-party tools, like CloudFoundry, Kubernetes and Mesosphere.
