The journey toward integrated management

To continue the thoughts of the previous newsletters, IT is currently faced with implementing management silos for each level of their infrastructure.  Or, perhaps silo is not the right analogy, as it would seem to slice vertically through the proverbial application data stack.  Rather, management today is like stratifications of sedimentary rock that may lie next to each other but have no mixing layers that herald the transitions.

The question is how do we move to a more integrated management environment where a customer could pick and choose from a menu of management support, depending on their needs? Of course, there are many answers to this question, some more complex than others. It seems to me that the place to start is with (surprise) a set of well-defined interfaces across the management disciplines.

The Distributed Management Task Force (DMTF) proposed a model for such interfaces with CIM/WEBM.  More recently, the storage industry has created the Bluefin specification, a first step in defining management interfaces for storage hardware components from the CIM/WEBM model.  But, where are the other disciplines such as applications management, systems management, network management or even content management?

When I ask both vendors and analysts in these other spaces this very question, I often get the response, “Well, that will NEVER happen!”  Why? Quite simply, legacy management infrastructures. “Too much time and money has already been spent on conformance with the current frameworks using SNMP,” was one vendor’s response. I, of course, had to ask if they thought that SNMP was capable of supporting extended management capabilities.  Their response was, “Of course not, that’s why we have APIs!” … Yeah, aren’t “standards” great.

The point is there is a huge amount of management software already in the market that will not change unless the vendors are driven to change by customers. How vendors are convinced is by projecting, that’s right, return on investment (ROI) for the effort. Either current customers must be willing to directly pay for the added flexibility and personnel cost savings gained with a “management by menu” capability and/or the vendor sees market growth opportunities, which would expand their customer base significantly, to offset the (re)development costs.

So, customers want visible and fast ROI on their management investments but the ability of management software to add significant value across the enterprise is diminishing with the current stratified environment.  Vendors want visible and appreciable ROI on any development that would support changing the environment to enable interfaces across these current stratifications. Seems like the customer and the vendor are between a rock and a hard place.