Indexing techniques for MIB considering the CMIS operations

In many studies it was proved that an object-oriented database (OODB) is appropriate for large management information base (MIB) implementation. On the assumption that MIB is implemented using OODB, it can be easily guessed that the query forms on the OODB would be influenced by the common management information service (CMIS) operations considering the fact that in the OSI network management, the MIB is managed through the common management information service element (CMISE). We classify the forms of the queries on MIB into eight types and propose four indexing techniques for the efficient CMIS operations. The CMIS operations involve scoping and filtering which are believed to be the most powerful functions but cause much overhead. For scoping or filtering, a CMISE user performing a CMIS operation should reference the management information tree (MIT) which represents the containment relationships among the managed objects. So, taking into account of the containment relationships we propose the class index, the MIT index and the level index for scoping and the attribute index for filtering. These indexes would be useful for the efficient implementation of the OSI network management system