San Francisco and Cologne, Germany – July 29, 2021 – ArangoDB, the leading open source multi-model graph database, today announced the GA release of ArangoDB 3.8. ArangoDB 3.8 includes new graph query and search functionality, helping to meet increasing demand for businesses to perform graph-powered analytics at scale.
According to Gartner, graph adoption will double by 2022 [1]. The research firm also predicts that graph technologies will be used in 80% of data and analytics innovations by 2025 [2]. ArangoDB 3.8 adds a number of new features for analytics at scale, combining graph, full-text, and Geo analytics into a highly-scalable database.
Key new features in ArangoDB 3.8 include window operations, allowing analytics for time-series or user-defined windows of data, as well as a new ‘weighted’ graph traversal. ArangoDB 3.8 also introduces geo-spatial search, composable Analyzers, and computed fields for ArangoSearch, ArangoDB’s natively-integrated, full-text search and ranking engine. This advanced search functionality further rounds out the database’s aim to provide the flexibility and scalability developers need to bring solutions to market as simply and efficiently as possible.
“Enabling data analytics and storage at scale – especially centered around graph – is one of the central use cases for many of our open-source users and customers,” said Jörg Schad, PhD, CTO at ArangoDB. “ArangoDB 3.8 is a particularly exciting step in this direction as it includes a number of features highly-requested by our community and customers.”
Window Operations
Window operations power aggregations over related rows, producing a result for each query row. By being able to query ‘moments in time’, window operations helps ArangoDB users perform time-series analytics over their data.
Weighted Traversals (optimized for ArangoDB Enterprise)
ArangoDB 3.8 supports a new graph traversal type, ‘weighted’, which enumerates paths by increasing weights. This enables queries such as calculating travel times between cities, helping to support use cases including turn-by-turn directions or mapping delivery routes. When used in an Enterprise Edition deployment of ArangoDB, especially combined with features like SmartGraphs that optimize graph queries at scale, the power of weighted traversals is enhanced across sharded data.
ArangoDB 3.8 sees the addition of new Analyzers to ArangoSearch. The pipeline Analyzer allows multiple Analyzers to be ‘chained’ together, where the output of an Analyzer is passed to the next for further processing. The pipeline analyzer is designed for advanced text search and analytics, for example first normalizing text for case insensitive search followed by n-gram tokenization.
The AQL Analyzer serves as a data transformation tool, as it is capable of running an AQL query to perform data manipulation and / or filtering. This further rounds out ArangoDB’s functionality as a multi-model database, helping businesses eliminate the need for ‘yet another tool’ to perform analysis on their data.
ArangoDB 3.8 also adds support in ArangoSearch for Geo-spatial queries with the introduction of two new Geo Analyzers, geojson and geopoint, which can break up GeoJSON and JSON objects into a set of indexable tokens for further use with ArangoSearch Geo functions. Additionally, ArangoDB 3.8 includes new ArangoSearch Geo functions which enable geo-spatial queries, such as contains, distance, in-range, and intersects, allowing the combination of full-text and geo search.
Other key features in ArangoDB 3.8 include:
- New metrics API and ‘out-of-the-box’ dashboards: ArangoDB 3.8 features a new metrics API that follows Prometheus conventions for metrics. The new metrics API can publish predefined dashboards to Grafana based on personas such as ‘operator’, ‘database administrator’, and ‘user’. There is also a dashboard with all metrics, sorted into categories. ArangoDB 3.8 supports over 200 metrics and nearly 300 graphs in the complete dashboard.
- Performance improvements: ArangoDB 3.8 brings performance improvements to shard synchronization which make initial shard replication and synchronization up to 10X faster. It also introduces a new default per-query limit to prevent rogue queries from consuming too much memory of an ArangoDB instance.
ArangoDB 3.8 is available immediately for download here, and will be coming soon to ArangoDB ArangoGraph, ArangoDB’s managed service.
Learn more
- To watch a webinar to learn more about how ArangoDB 3.8 supports analytics at scale: Register here
- To get started with ArangoDB ArangoGraph: Sign up for a free trial
