CA puts a stake in the ground

At last week’s NetWorld+Interop 2003 trade show Computer Associates introduced its “Managing On-Demand Computing” initiative, which is corporatewide – addressing storage, security and e-business – and uses Unicenter as its management lynchpin.

In spite of the general industry buzz already surrounding the need for dynamic, on-demand management, CA has made this announcement distinctive and compelling in a number of ways. For one, it has united a well-defined architectural structure with a set of human/organizational requirements. It has also made commitments for significantly enhanced analytics (including root-cause analysis) and more flexible integration – both of which have been areas of weakness for CA in the past. And it has enhanced an already advanced strategy for visualization by establishing a new product family of consoles for role-based information and services.

The “Managing On-Demand Computing” initiative is rich in concepts, not all of which are self-evident. CA introduced three criteria, or objectives, around which to design and develop its products and architecture. These are:

* Delivering IT as a service, or enabling business-relevant capabilities so that IT can function as an integrated part of the business.

* Self-management, with a focus on automated capabilities to ensure self-healing and self-adaptive infrastructures.

* Service-oriented architecture, which is about enabling role-based services and information across products from CA and third parties.

Just as importantly, CA has also defined three contexts or markets for delivering products:

* IT Operations – the day-to-day business of ensuring that the infrastructure performs as it must.

* Service – the mixture of customer interaction and service planning that cuts across help desks, operations centers and business planning today and helps to ensure that business and service obligations are fully met.

* IT Resource – a renewed focus on asset management and support that over time will include the full networked environment as an investment to be dynamically measured and accounted for. Ultimately, “resources” include service provider services and switches on a port, just as they must include PCs and servers.

While going into any architectural detail is beyond the scope of this column, suffice it to say that CA is making significant investments in analytics. For instance, CA will introduce in coming months its Distributed Intelligent Architecture, with analytic DIA agents, and capabilities to support desired-state management and self-healing.

Perhaps even more fundamental is CA’s ongoing investment in Common Services, for secure data transport across the CA portfolio. Its support for compound objects in the Common Object Repository will support a tiered approach to analytics, as well as a profound investment in visualization and service access that can bring, for instance, asset, performance, and topological data together in a common context.

From a product introduction perspective, I would like to single out CA’s family of “Role-based Consoles” as emblematic of this initiative. Specific to this announcement are three console groups (CA foresees variations that could extend consoles to more than 50 user types): Business User, IT Operations User and Service Management.

Business user consoles will support such views and functions as cost information; service-level agreements; procurement support; resource allocation and usage; dashboards reflecting overall infrastructure health, ROI, NPV and other metrics; and reports justifying new investments or scoring service provider performance.

In contrast, the IT Operations User suite of consoles (for users ranging from IT directors to network operators and desktop managers) will provide real-time GUIs targeting brownouts and blackouts, map trouble tickets to service contracts, and provide consistent data to support collective problem solving rather than finger-pointing across networks, systems and applications specialists. It will also have optimization reports, along with other, more specialized functions. 

The idea of clustering function based on role and decision requirements, as well as creating deliberate capabilities to build bridges across IT silos, is both enlightened and ambitious. This is not “just another rollout” in my view, but instead a real stake in the ground expressing a no-holds-barred commitment to evolve towards a “next-generation” design that’s modular, automated and sensitive to business, organizational and personal requirements. It will be worth watching CA in the coming months and years – as it steps up to answering the full reach and scope of its new vision.