Who we are

LSYSTAP offers a variety of consulting services around Blazegraph and MapGraph, including application-specific data access layer integrations, performance tuning and query optimization, custom feature development, and on-site training. Many of our most popular features have been underwritten by enterprise customers with a particular need. Please contact us directly for more information.

SYSTAP coined and trademarked the term “bigdata” in 2006 as a name for our open-source graph database platform. The name and concept for Bigdata was derived from Google’s 2006 paper on BigTable.

In 2015, SYSTAP launched the Blazegraph&tm; graph database built on Bigdata® with 100% binary compatibility.

Bigdata® is a registered trademark of SYSTAP, Blazegraph(tm) is a pending trademark for SYSTAP.Mapgraph(tm) is a pending trademark for SYSTAP.LLC. SYSTAP takes great care in the development and protection of its trademarks and reserves all rights of ownership of its trademarks. No other company may use SYSTAP’s trademarks unless it has the express written permission of SYSTAP, or is licensed by SYSTAP to do so.

licensed by SYSTAP

Graphs, RDF, And Blazegraph

Graphs are a powerful, flexible means of representing all kinds of linked data. RDF is a data model that provides a standards-based way of describing, interchanging, and querying graph data. RDF models a graph as a collection of RDF statements. RDF statements let you add vertices, edges, and vertex properties to your graph. RDR is a simple extension to RDF that allows you to add edge properties as well.

Blazegraph is a fully open-source high-performance graph database supporting the RDF data model and RDR. Blazegraph operates as an embedded database or over a client/server REST API. Blazegraph supports high-availability and dynamic sharding. Blazegraph supports both the Blueprints and Sesame APIs.


What Is RDF?

RDF is most often associated with ontologies and semantics, but it is also the only standards-based means of describing, interchanging, and querying graph data. RDF is a schema-flexible graph data model. RDF Schema defines a simple framework for doing basic type classification (class and property hierarchies). OWL allows you to define rich ontologies and perform complex reasoning tasks over your data. This makes RDF an excellent choice for data fusion – bringing together disparate data from many sources in many formats and and then using the constructs in RDFS and OWL to do real-time semantic alignment. RDF also offers the only standards-based high-level query language for graph data – SPARQL, a powerful query language supporting update, complex joins, nested subqueries, aggregations, federation, custom functions and services, and much more. RDF models a graph as a collection of RDF statements, defined by a subject, predicate, and object. A statement can be used to define an edge between two vertices, or to define a property on a vertex (or edge). SPARQL allows you to select subsets of your data by defining statement patterns consisting of variables and constants. Statement patterns that share a variable constitute a join. # create a graph with four vertices and three edges ex:alice foaf:knows ex:bob . ex:bob foaf:knows ex:mary . ex:sue foaf:knows ex:mary .     # attach some attributes to the vertices ex:alice rdfs:label “Alice” . ex:bob rdfs:label “Bob” . ex:mary rdfs:label “Mary” . ex:sue rdfs:label “Sue” .   # find me everyone who knows Mary select ?who ?label where { ?who foaf:knows ex:mary . ?who rdfs:label ?label . } # Bob and Sue both know Mary results: [ (?who=ex:bob, ?label=”Bob”), (?who=ex:sue, ?label=”Sue”) ]


Scaling Your Graph

Blazegraph scales to 50 billion (50,000 million) edges on a single commodity server. Blazegraph supports optional transactions, very high concurrency, and very high aggregate IO rates. Blazegraph was also designed from the ground up to be capable of running as a distributed database architecture across clusters of 100s or 1000s of machines. This horizontally distributed architecture shards and load balances your graph data automatically. Contact us to schedule a time to discuss the specifics of this architecture and how you can start using it.


The Big Graph Anti-Pattern

Many graph database and graph analytics systems wrap key stores with a graph layer, which facilitates what we like to call the “Big Graph Anti-Pattern” – throw your Big Graph into a key-store to get it sharded and call it a day. But this approach falls over at scale for graph query and graph mining/traversal, because simple data parallel abstractions fail to recognize several important differences with graphs. Read more about the Big Graph Anti-Pattern here.


Data Interchange In RDF

RDF provides you with a standardized means of interchanging graph data in a number of different formats. Learn about RDF/XML, Turtle, and some of the other interchange formats here.


High-Level Query With SPARQL

SPARQL is a powerful W3C standard for high-level query of graph data. SPARQL is a powerful query language supporting update, complex joins, nested subqueries, aggregations, federation, custom functions and services, and much more. Learn more about SPARQL.


Inference, Type Classification, And Rules

RDF Schema and OWL provide for rich type classification and inference, and Blazegraph gives you a pluggable way to define custom rules for your application. The BigdataSailFactory has an Inference option that will enable basic RDFS and OWL inferencing at load time. Blazegraph supports: RDFS Classification rdfs:domain, rdfs:range, rdfs:subClassOf, rdfs:subPropertyOf OWL type and instance equivalence owl:equivalentClass, owl:equivalentProperty, owl:sameAs Other OWL constructs owl:FunctionalProperty, owl:InverseFunctionalProperty, owl:SymmetricProperty, owl:TransitiveProperty, owl:inverseOf, owl:hasValue You can also turn these constructs on and off individually, or create custom inference rules of your own with Blazegraph’s pluggable rules engine. Consult the Wiki and documentation for more details. Blazegraph supports incremental truth maintenance for insert and delete. Be aware that incremental truth maintenance runs by default, and is not suitable for bulk load operations. To bulk load data into a Blazegraph instance with Inference enabled, use the high-performance data loader: final BigdataSailRepository repo= BigdataSailFactory.createRepository(“/tmp/bigdata.jnl”, Option.Inference);   // bulk load data without incremental inference final DataLoader loader = repo.getDatabase().getDataLoader(); …


Edge Attributes In RDF

RDR (“Reification Done Right”) is a new specification for asserting statements about statements in RDF. RDR lets you treat a statement as an RDF Resource that you can then use inside of another statement as a subject or object. Although RDR is not an official part of the RDF data model yet, Blazegraph supports RDR for triples, and will support RDR for quads in an upcoming release. Currently, only Blazegraph’s Turtle parser and Turtle writer support the RDR syntax: # assert some ground triples ex:alice rdf:type foaf:Person ; foaf:name “Alice” ; foaf:mbox <mailto:alice@work> ; foaf:knows ex:bob . # assert triples about triples with the RDR <<>> syntax << ex:alice foaf:knows ex:bob >> dc:source <http://hr.example.com/employees#alice> ; dc:created “2012-02-05T12:34:00Z”^^xsd:dateTime .   The RDR syntax is also built into Blazegraph’s SPARQL grammar, allowing you to query using the same syntax: # query for all the assertions about a known statement select ?p ?o where { << ex:alice foaf:knows ex:bob >> ?p ?o . } # or query for ground statements based on RDR assertions select ?s ?p ?o where { << ?s ?p ?o >> dc:source <http://hr.example.com/employees#alice> . } Learn more about the new RDR syntax here and here.


Graphs, RDF, And Blazegraph

Graphs are a powerful, flexible means of representing all kinds of linked data. RDF is a data model that provides a standards-based way of describing, interchanging, and querying graph data. RDF models a graph as a collection of RDF statements. RDF statements let you add vertices, edges, and vertex properties to your graph. RDR is a simple extension to RDF that allows you to add edge properties as well.

Blazegraph is a fully open-source high-performance graph database supporting the RDF data model and RDR. Blazegraph operates as an embedded database or over a client/server REST API. Blazegraph supports high-availability and dynamic sharding. Blazegraph supports both the Blueprints and Sesame APIs.


Companies that chose Blazegraph

The gallery was not found!