Difference between revisions of "Bigdata"

From Semantic Web Standards
Jump to: navigation, search
Line 6: Line 6:
  | SWTechnology3=RDFS
  | SWTechnology3=RDFS
  | SWTechnology4=OWL
  | SWTechnology4=OWL
| SWTechnology4=Blueprints
  | Category1=Triple Store  
  | Category1=Triple Store  
  | Category2=Reasoner  
  | Category2=Reasoner  

Revision as of 12:04, 19 June 2014


Name of the tool: Bigdata®
Home page: http://www.bigdata.com/blog
Date of latest release:
Programming language(s) that can be used with the tool:
Relevant semantic web technologies: RDF, SPARQL, RDFS, OWL
Categories: Triple Store, Reasoner, RDFS Reasoner
See also: http://www.mapgraph.io
Public mailing list:
Preferred project URI:
DOAP reference:
Company or institution: SYSTAP, LLC.

(Tool description last modified on 2014-06-19.)


Bigdata® is a standards-based, high-performance, scalable, open-source graph database. Written entirely in Java, the platform supports the RDF data model and the SPARQL 1.1 family of specifications, including Query, Update, Basic Federated Query, and Service Description. Bigdata can be used for triples or quads and supports RDFS+ inference. Bigdata supports novel extensions for durable named solution sets, efficient storage and querying of reified statement models based on <a href="http://arxiv.org/abs/1406.3399">RDF* and SPARQL*</a>, scalable graph analytics using the vertex-centric Gather Apply Scatter (GAS) abstraction, and interoperability with the Blueprints API. The database supports multi-tenancy and can be deployed as an embedded database, a standalone server, a highly available replication cluster, and as a horizontally-sharded federation of services similar to Google’s bigtable, Apache Accumulo, or Cassandra.

Bigdata is a highly scalable platform. The standalone configuration of the database scales up to 50 billion (5x10^10) triples or quads on a single commodity server with a sustained bulk insert rate of 50k-80k statements per second.

Bigdata 1.3.0 introduces support for highly available replication clusters with low-level write replication, linear scaling in query, automatic failover, self-healing, online backups, and more.

The bigdata open source platform has been under continuous development since 2006. It is available under a dual licensing model (GPLv2 and commercial licensing) and a number of well-known companies OEM, resell, or embed bigdata in their applications. SYSTAP, LLC leads the development of the open-source project and offers support subscriptions for both commercial and open-source users. Our goal is a robust, scalable, high-performance, and innovative platform.

The related MapGraph (http://mapgraph.io) platform supports ultra-high performance parallel graph algorithms at up to 3,000,000,000 Traversed Edges Per Second (TEPS) on NVIDIA GPUs.

(Open source. Developer Support, production support, and commercial licensing available on request.)