Databases

    From Consumerium development wiki R&D Wiki
    Revision as of 20:15, 26 August 2016 by Jukeboksi (talk | contribs) (→‎Object database: more info from Wikipedia)

    This article is about choice of database models and implementations.

    Copyleft free to modify and free in cost software is strongly preferred over other solutions. Minimal modifications required could be another preference as that means maintenance of the chosen solution is minimized in that aspect.

    Known types of databases

    • Relational database provided by a RDBMS and queried with SQL. Track-proven technology.
    • w:NoSQL variates:
      • w:Subject-predicate-object databases are implemented by graph databases, specialized native triplestores and piggy-packing solutions that use an RDBMS to store and query the triplets and the networks they compose.
      • Graph databases would intuitively appear more advanced than using RDF-triplet composed semantic networks but are not much different on the outside. Both jump through the same hoops but with different efficiency and grace.
      • Object databases are old but on the rise with NoSQL-based thinking and the modern needs, like leanness, real-time need and scaleability for which the other solutions might be too limiting.

    Relational database

    Relational databases work by storing data in tabular form where columns represent data items of predetermined type and rows represent the values each "item" has. Relational databases are accessed mainly with SQL ( Structured Query Language ). However the RDBMS converts that into relational algebra and optimizes that and the relational algebra query actually returns the result table that has those columns and rows you requested.

    Together Consumerium and Consumium run all the 3 major free full fledged RDBMS:


    NoSQL databases

    “A NoSQL (originally referring to "non SQL" or "non relational") database provides a mechanism for storage and retrieval of data which is modeled in means other than the tabular relations used in relational databases.”

    All the following database types can be considered variations of NoSQL.


    Subject-predicate-object database

    Subject-predicate-object databases basically construct w:semantic networks from interlinked atomic units called a w:triplet so they are not fundamentally different from graph databases in functionality and utility offered.

    These networks may be queried with a suitable query language such as w:SPARQL which in practice allows you to compose semantic queries.

    SPARQL is a recursive acronym and stands for SPARQL Protocol and RDF Query Language). It is an RDF query language, that is, a semantic query language for databases, able to retrieve and manipulate data stored in Resource Description Framework (RDF) format.”

    “A triplestore or RDF store is a purpose-built database for the storage and retrieval of triples through semantic queries.”

    Relevant subject-predicate-object database powered systems to interoperate with

    Things to consider in selection of triplestore

    “Some subject-predicate-object databases (also known as triplestores) have been built as database engines from scratch, while others have been built on top of existing commercial relational database engines (e.g., SQL-based).”


    A w:triplestore maybe a native implementation from ground up or be standing on the shoulders of a standard RDBMS system where actual w:SQL is formulated by the interpreter and then queried from SQL. This probably has upsides and downsides.

    Lists and comparisons of subject-predicate-object databases and SPARQL implementations


    Graph database

    A graph database stores and queries graphs.

    These graphs may be stored in and constructed from RDF triplets readily so they are quite alike and overlapping in functionality offered but the query performance varies (see talk page for more).

    Lists of graph databases

    Free reading on graph databases


    Object database

    “An object database stores complex data and relationships between data directly, without mapping to relational rows and columns, and this makes them suitable for applications dealing with very complex data.”