The multi-model Big Graph Store
A massively scalable Big Data graph store designed for holding and querying vast amounts of raw data.
It can store objects and links with multi-valued properties and raw data, fast queries, aggregations, distributed data processing pipelines and super-strong security (row and property-level).
Can be used to build solutions that cover most data analytics use cases and industry requirements, from financial analysis to social media and cybersecurity.
Can be deployed on-premise, in the cloud, or both.
- Dynamic data model (concepts, relations, properties, tables), multivalued properties and property metadata support
- Authorizations and visibilities for rows and columns
- Indexing for fast search: full-text, fuzzy, spatial, aggregations
- Async distributed processing pipeline for applying custom logic on import/update/delete operationg
- Cypher query language support
- Super extensible and modular
- Beautiful feature-rich visual explorer
- Developed in Java
Have a look at the Architecture and Concepts page to understand how the platform works.
- Memory
- RocksDB
- Apache Accumulo
- ElasticSearch
The visual face of BigConnect, available in the bigconnect-explorer repository.
- Dashboard with configurable widgets
- Collaboration using spaces
- Global faceted search with aggregation support
- Graph visualization with advanced features
- Charts
- Map visualization
- Timeline visualization
- Custom item viewer for documents, video, audio & images
- Annotation of text & images
- Data Schema configuration
- Extensible at all levels with plugins
- LDAP & SSO integration
- Developed in Java, React & FlightJS
BigConnect can be run just about anywhere using the deployment methods described below.
To run the latest build via Docker with the RocksDB backend and embeded ElasticSearch, just type
docker run -d -p 10242:10242 -p 10243:10243 bigconnect/bigconnect
If you would like to persist data between runs you can supply a volume to the docker run command:
docker run -d -p 10242:10242 -p 10243:10243 -v /tmp/datastore:/bc/datastore bigconnect/bigconnect
Now, open a browser and go to http://localhost:10243/cypherlab/index.html
Login with the following details:
- Connect URL: bolt://localhost:10242
- Username: admin
- Password admin
The easiest way to interact with BigConnect is to use the Cypher query language.
Get started by following the Cypher tutorials available here.
To run the latest build you need to build the project first. You need to have Maven installed and JDK11.
mvn -Pbin-release clean install
The release artifacts are available in the release/target/bc-core-*
folder
Advanced installations require a Hadoop cluster with Apache Accumulo 1.9, RabbitMQ and an ElasticSearch 7.6+ cluster.
Contributions are warmly welcomed and greatly appreciated. Here are a few ways you can contribute:
- Start by fixing some issues
- Submit a Pull Request with your fix
BigConnect is an open source product licensed under AGPLv3.