Every now and again it's healthy to rise above the hottest new storage technologies and try to gain a broader perspective. I like to call this perspective "Business Value."
To quote Martin Luther King, "I have a dream!" My dream, similar to Dr. King's, is for integration; integrated management of all assets across the business infrastructure.
There are others out there with the same dream. CIOs struggle daily with having to spend their IT budget dollars on chewing gum and bailing wire - or the technological equivalents - to keep going the infrastructure that runs the business. Those of you who just spent hundreds of thousands of dollars to implement the latest and greatest management software, which now needs significant scripting to integrate with the business' current best practices, know exactly what I am talking about.
But let me tell you a little more about my dream. In my dream, all IT assets, both hardware and software, provide services.
These services can be combined in a number of different ways to create a service chain that supports business process.
Embedded, as part of these services (yes, this should be a given), comes the management interfaces. These interfaces describe the asset, rules for managing the asset and provide continuous health information about the asset.
Because business requirements change, the service chain must be flexible enough to change as well. Of course, this means the services' management interfaces must be equally flexible; even (do I dare say it?) plug-and-play. To accomplish this, each service type must conform to a standard model with consistent interfaces.
I can hear the screams now. "But standardizing the management of all like-services would only provide the lowest common capabilities for management!" Here I disagree and it is where I think many technology companies are shortsighted (remember to look up sometimes). When you view every asset as providing a service, then it is easy to describe and build service chains for different levels of service.
As an example, let's look at a CIO who is implementing a customer resource management system. This system will track a customer's contact and configuration information, sales and service history, and order fulfillment. This system touches the heart of the company's revenue stream. If it is not available, it costs the company money, in time and possibly lost sales.
This business process requires extremely high service levels.
As such, the service chain that would be built to support this process would have to provide the required services levels.
Therefore, each of the assets that make up the service chain would, individually, have to provide a level of service to meet the quality of service required.
In the storage domain, this may mean using high-end, high-performing storage subsystems that allow for automatic snap-shots of the data for replication to another building or another site, combined with redundant infrastructure, automatic failover, role-based security, etc. The list goes on and on.
But what happens if the service chain needs to be reconfigured for some reason? Perhaps the storage subsystem is not performing fast enough, perhaps an infrastructure component won't scale as needed? Today, managing this level of change is excruciatingly painful. In a service-based world with standard interfaces, managing this type of change can be simplified (of course, this applies to all components within the infrastructure including servers, networks, underlying applications such as databases and backup/recovery software...).
As you can see, in this dream, IT infrastructure would be managed with regards to the business process(s) rather than the technology. One group that is trying to make this dream a reality is the DMTF.
DMTF used to stand for "Desktop Management Task Force" and was recently renamed "Distributed Management Task Force." The DMTF is made up of industry vendor companies, other industry associations (like the Storage Network Industry Association), academics and end-users... some of the brightest minds in management. The mission of the DMTF is "To lead the development of management standards for distributed desktop, network, enterprise and Internet environments."
While we have not realized the dream, yet, accomplishments like the recent Bluefin specification that came out of the SNIA are the stepping stones down the path to pervasive, service-based, plug-and-play management of the entire IT infrastructure