RDF-based IT Configuration Management Database

Introduction

The Information Technology Infrastructure Library (ITIL) is a set of concepts and practices for managing Information Technology (IT) services (ITSM), IT development and IT operations.
Nowadays, ITIL is one of the most respected and prestigious set of best practices for IT infrastructures, having many research papers and products (e.g., HP Openview, IBM Tivoli) that are based on.

In ITIL's context, the library recommends the use of a Configuration Management Database (CMDB). Basically, the CMDB is responsible to store an updated and complete view of the IT infrastructure configuration. Some examples on this database are: the server X has 21Gb of free memory, the switch Y is capable to transfer 1 GB/s, the switch Y is located at room 10, the server W is connected through a VLAN to server X, and others. In fact, all the relations between IT resources and services should be reflected at the CMDB in an updated manner.

In order to build such CMDB, a powerful model is required. One of the most well-known models for this purpose is the Common Information Model (CIM). TheCIM is an open standard that defines how managed elements in an IT environment are represented as a common set of objects and relationships between them. This is intended to allow consistent management of these managed elements, independent of their manufacturer or provider.

Problem

Most implementations of CMDB are:

• Based on a non-standardized model, but are open source;
• Or follow the CIM model, but are not open source.

Another eminent problem is that once the CMDB is populated respecting a model, it is difficult to reasoning and infer information over the data. Example: which server has X, Y, and Z features, and have the fastest link/connection to server S?

Theses Description

This theses aims to design, develop, and publish an open source RDF-based IT Configuration Management Database (CMDB) following the CIM model for IT management purposes.

The Resource Description Framework (RDF) is is similar to classic conceptual modeling approaches such as Entity-Relationship or Class diagrams, and it is based upon the idea of making statements about resources using the form of subject-predicate-object expressions (triples). Therefore, expressing the CIM model based on RDF will enable a more organized and practical manner to persist and reasoning/infer IT infrastructure information.

Some tasks for this theses include (in high-level):

• Map a subset of CIM from UML to RDFS (Resource Description Framework Schema)
• Integrate the resultant schema (which is actually the CIM model) to a RDF database (e.g., Sesame)
• Build a library to manage some elements in an easy way (interface)
• Pack the solution into a usable and generic manner to be published