Knowledge is profit!


Brief API description

API demo

Performance benchmark

ArchiGraph.MDMsemantic MDM solution

The purpose and principle of semantic MDM

MDM (Master Data Management) systems are intended for storing and managing the reference data set. These data are used by all the other applications in the enterprise. MDM may host catalogues of the customers, assets, employees, shared classifiers and enumerations, and many more.

A huge variety of the structure of the master data hardly fits into relational database model. It is common that entities of the same type may have variable attributes set depending on their classifications, etc. This leads to an idea to use semantic technologies to represent and manage master data. This idea is implemented in our ArchiGraph.MDM.

A semantic MDM system has the next advantages:

  • Semantic MDM is able to store data having unlimited complexity of the structure, containing tens and hundreds thousands of entity types. It allows creating subject matter area descriptions without excessive simplification, without loss of expressivity and meaning.
  • As in the semantic model the data structure is not separated technically from the data itself, it is much easier to manipulate them. Such system is able to react most effectively on any changes of business requirements, leading to changes of data model.
  • Semantic MDM offers programmatic access to the data model to the client applications.
  • Semantic MDM can be conveniently used for storing application components settings and work algorithms, thus facilitating their governance. This offers possibility to develop model-driven software, which is much easier and cheaper in support.

ArchiGraph.MDM functionality

Our product performs the following functions:

  • Storing data model, including definitions of classes, possible attributes and relations of their instances, providing ability to make any particular object a member of any number of classes, to organize classes into several hierarchies.
  • Indicating constraints on attribute values, including minimal and maximal number of possible values for each attribute, domains and value ranges.
  • Storing information on particular objects, their classifications, attribute values and relations. Maintaining unique identifiers of the objects.
  • Providing programmatic access to the data model structure.
  • Restricting access level for the client applications according to the access rights settings defined for classes/applications.
  • MDM provides logging of all the queries processed, and gives an interface to surf these logs.
  • MDM can work in cluster mode, providing extra reliability in query processing.

For data model construction and maintenance, a special editor is used. It has the next functions and features:

  • Multiuser access. Access can be granted to the users on the classes level. The set of operations (no access/see/change) available for the user with the particular entity will be determined as the most strict access level defined or inherited for all the classes with which this entity is associated.
  • The editor allows to create, edit, delete model elements of every type, view them using trees or lists with sorting and filtering, perform fast search, assign values for the attributes of arbitrary groups of entities (including assigning several values for each attribute), add user comments and attach files to the model elements.
  • Editor supports importing/exporting model parts using Excel files.
  • Editor tracks history of model editing and provide access to the history for the users.