Geek Answers Handbook

What are the strengths and weaknesses of the tripestor database?

What are the strengths and weaknesses of a triplestore database like Mulgara ? In what situations would its performance differ significantly from that of a traditional relational database?

I think in particular about situations where you have data in the form of namespace:predicate=value . Triplestores seems to store data in its “natural” form, but I would like to know if there are any additional benefits beyond this.

+6
performance database rdf triplestore
source share
2 answers

I assume that you are asking about storing a triple (and possibly quads), where the store is either a “native” triple store (like TDB ) or a relational database with triple store (like SDB ).

If you understand your requirements in detail, I am sure that the members of Qaru can offer a more direct answer, but only from my head:

  • Native triple stores usually have better performance when the data set size becomes very large. See some numbers from the W3C .
  • Native triple stores may offer in-store reasoning / inference (e.g., Oracle semantic tools, although it does not fully support SPARQL).
  • The triples in the DBMS can be “easily” integrated into traditional reporting tools (for example, Jasper reports), since you can query them using SQL.
+2
source share

I am the creator of Redland , which includes its own thrice stores in C plus. I used some of Java such as Jena and Sesame.

It's hard to say more without knowing what the problem is, but in general, native triple stores work better if your problematic space / request template is more “graphic” rather than “objective”. An object means that you tend to always manipulate - set / receive the same sample of data each time - in this case you can use ORM or another object-relational mapping and a fixed schema. If your problem and scheme is more dynamic - "Graphy" - you can take a triple approach to storage, where it’s easier to jump over the data than usual in the aka join world.

Of course, there is also an approach to the document, such as CouchDB, but which is difficult to learn from the information you provided.

+5
source share

More articles:


All Articles