CODEWITHSUNDEEP
rdfjenatriplestore
S M Shamimul Hasan
S M Shamimul Hasan

Reputation: 6654

Efficient Storage and Querying of RDF triples

For my project I have a relational DB. Frequently my relational DB schema is changing. So recently I stored all the relational database tables data into one big RDF table. Table stored data as a triple. It has three columns Subject, Predicate, Object. Following is an example of the RDF Table:

Subject->Predicate->Object 

1->name->Center

1->description->sample description

1->data->measure

2->data->parameter

2->relation->1

I stored the table in Oracle and currently using SQL to query the DB. However, my table is growing very fast and it has lots of records and SQL query processing taking time. I believe in the near future table will grow more and query processing will be inefficient. So, my concern is about efficiency.

Could you please inform me how can I store RDF triples efficiently so that it will take less time to query the DB even though the table is very large.

I don’t know much about RDF database and query language. Is it possible to make Oracle RDF table efficient? I saw that people are talking about Apache Jena tool and SPARQL language. Will it solve my problem? If I use Jena then do I need to store my RDF triple in XML format or my current format is fine? How I can use Jena tool? Also will it be helpful to use graph database instead of Oracle DB? If so then which one to choose?

Upvotes: 4

Views: 904

Answers (1)

AndyS
AndyS

Reputation: 16630

http://jena.apache.org/

Jena has two storage layers that apply here : SDB, which uses SQL databases and TDB, a native store. TDB is faster. SDB uses it's own schema to store RDF.

You can also Oracle's won RDf storage (which has a jena interface).

You do not store XML for the RDf in any of these approaches - it is parsed and efficient storage used.

Upvotes: 2

Related Questions