Next generation management system structure begins to take shape
|
|
|||
|
|
Sign up to receive this and other networking newsletters in your inbox.
However, given the pace of change in Internet time and the fact that few, if any, management responsibilities are ever really taken away, the task at hand is about to expand as well as the protocol and information hierarchy designed to support it.
A recent press release from the Distributed Management Task Force (DMTF) described its plans to release a proposed standard in June for using HTTP to exchange system and network management data over the Web. This effort enhances, and will encompass, the work originally begun by the Web Based Enterprise Management consortium, in the areas of protocol definition, information models and the use of directory services.
In addition, an information model is being defined that captures notions that are common to particular management areas but independent of a particular technology or implementation. The common areas are systems, applications, databases, networks and devices. The information model is specific enough to provide a basis for the development of management applications. This schema provides a set of base classes for extension into the area of technology specific schemas. There are currently five Common Schema areas:
- Systems
- Applications
- Networks (LAN)
- Devices
- Physical
Users should care about this initiative because of the functional scope of coverage it provides as well as how it continues the work begun by players such as BMC Software, Cisco, Compaq, Intel, and Microsoft and pulls in contributions from Dell, Novell, IBM/Tivoli and Sun.
This work will carry forward the use of Extensible Markup Language (XML) to add significant management value that really could not be easily supported by SNMP. XML has significant potential as a data description language that will let a web browser and a server exchange a richer set of tagged data that cannot be easily supported using standard HTTP. XML applications are most significantly used in exchanging management data between otherwise incompatible element managers and describing the structure of nonmanagement entities such as electronic transactions that can, in turn, be more effectively understood and controlled by next generation management products.
The real value that this effort brings, in my opinion, is the means to bring together management of three separate domains - traditional enterprise management, electronic business service management and service provider service management - into a more consolidated interface and data definition hierarchy with the browser as the universal access mechanism. The enterprise manager of the future will need to be concerned about managing all three domains. Without some form of industry-supported architecture to standardize the interface, data definition standards and data definition rules, this management task is hopeless.
Plans for these projects expect to be rolled out at the DMTF meeting in San Jose, June 15 and June 16. Prototype products based on the initiatives are planned for demonstration at the meeting. Users can expect some initial prototypes and perhaps even a few initial products to be demonstrated at NetWorld+Interop in Atlanta.
