An api for domain level interaction with SNOMED CT and an implementation based on a storage engine using a native SNOMED distribution stored in neo4j as a backend.
- Anders Nawroth
"Typical strenghts of a graph database are handling of complex relationships, social network analysis, path finding, deep traversals, minimal gap to the domain model. Weaknesses are lack in tools and reporting. Neo4j is primarily exposed as an embedded DB through a Java API, but any JVM language could make use of it. Currently there exist bindings for Python, Ruby, Scala and Clojure. There's also support for domain-centric REST APIs on top of the Ruby and Scala bindings, using JSON for the data. To avoid vendor lock-in you could use the graph data model approach provided by the RDF/SPARQL standards. Neo4j supports this as well. AllegroGraph RDFStore is another graphdb, with client libraries in C#, Lisp, Java and Python. There's REST support as well. Disclosure: I'm on the Neo4j team."
- Anders Nawroth
AliBaba is a collection of modules that provide simplified RDF store abstractions to accelerate development and facilitate application maintenance.
- Anders Nawroth
An overview of the last couple months' debate in the community about various emerging database paradigms. The post summarizes arguments against the relational model, advantages with key-value stores and relates them to graph databases.
- Anders Nawroth
"Thx a lot for sharing this! To me this looks like part of the tendency to move away from RDBMS for some or most of the storage needs to be able to scale well. The next natural step would be a shared nothing architecture using specialized storage engines for different needs. For instance there are different key/value stores with varying characteristics and the document-based ones should get more stable over time. One option I think will get more and more interesting in the future is using a graph database engine like http://neo4j.org/ (which BTW is the reason I joined that project) - it handles interconnected and semistructured data in a very efficient way."
- Anders Nawroth