A group of vendors Monday announced a draft specification that industry watchers say could enable information sharing among multiple configuration management databases and other management products, such as inventory tools and service desk systems.

Companies such as BMC, CA, HP and IBM have been for the past few years pitching their respective CMDB tools. As members of the CMDB Federation working group, however, these management heavyweights, along with Fujitsu and Microsoft, have drafted a specification designed to provide standard ways for data sharing among multiple configuration products. So far it remains unclear with which standards body the specification will land for further development.

“The CMDBf has not finalized which standards body the specification will be presented to for standards consideration. There are several organizations that are possible, but it is premature to speculate which standards body the working group will approach," says Marvin Waschke, CA senior technology strategist and an author of the draft specification.

According to the organization's Web site, the CMDBf Public Interim Draft version 0.95 defines how federated CMDBs and other management data repositories, or MDRs, can interact to share data. Specifically the draft "defines query and registration Web services" for the interactions between a federating CMDB and an MDR, and it is based on HTTP, SOAP, WSDL, XML Schema and Web Services Interoperability (WS-I).

Federated CMDBs let customers collect data from multiple sources without having to store all the data in a single, monolithic database. In the federated scenario, configuration data can reside in multiple sources and one centralized data sources is programmed to know where data related to specific assets resides. That way, IT managers don't have to become overwhelmed with the maintenance of one large data repository of data, industry watchers say.

"It tackles data access across multiple brands in a feasible and pragmatic way," says Dennis Drogseth, a vice president with research firm Enterprise Management Associates. "It doesn't tackle data reconciliation or normalization issues, which are tougher, but you have to run before you can run. It's a very good step forward, not a giant step, but a good one."