Temenos T24 Database structure

9.9k views Asked by At

I was working with R09 of Temenos T24 which had Oracle as the background. Table structure was 2 columns - RECID + Data in Blob (XML format).

Has anyone got an idea, if the structure has been changed to RDBMS structure with the new T24 versions such as R17 or R18?

Thank you for any help in advance !!!

4

There are 4 answers

0
Maratto On BEST ANSWER

Working with R15 - still RECID + Blob. I'm quite sure that R18 is the same as we're currently upgrading to R18 and no DB scheme change is in the road map.

0
aldiyudha On

you can select table view directly from DB, like SELECT * FROM V_FXXX_ACCOUNT. That table RDBMS you can select field that you need.

0
kadar On

Temenos do have a product called Relational Replication aimed at providing a selected table from T24 in a relational format. So all the multi-value / group multi-value elements become child tables and sub-value elements come in further child tables with foregin keys. So easier to index and query. They also have a data model viewer for T24 in Design Studio which gives you an idea of how these tables will be structured.

0
PrecisionLex On

Temenos T24 core was built around the so called "Multi Value Database" UniVerse and then moved to jBASE around year 2003. See this link https://en.wikipedia.org/wiki/MultiValue for an explanation what is a Multi Value Database.

Later, to add support for Oracle and other industry standard "big" databases, Temenos developed a special DB driver for their system that was designed to imitate the Multi-value database functionality inside the RDBMS. The solution was to use XML to store the multi dimensioned fields. And so all T24 tables in Oracle have two columns:

  • RECID for the ID or Unique Key of the Record
  • XMLRECORD to store the data.

The XMLRECORD by default is created as XMLTYPE, but can also be BLOB or CLOB type. In this case the data will be stored as it used to be stored inside the old Multi Value Database, i.e. a string where fields are separated by field markers, value markers and sub-value markers.

This basically means that T24 will never move to proper RDBMS structure as that would mean to completely re-write the whole T24 solution, or at least a significant part of it. Since T24 is being developed for 30 or more years now, you can imagine what it would take to perform such a task.