Managing What You Can't See: Service Management in a Virtual World

The service management world of  today is all about linking business services to the underlying IT infrastructure, creating an effective bridge between the business and technology. In theory, this provides a clear window into the IT environment to increase accountability, productivity and efficiency. Effective service management also provides business context, so IT can take action to avert service-impacting events by understanding business priority.

However, current business service management (BSM) does not provide enough guidance about how to manage services proactively and effectively. This issue is now more important than ever, because on the horizon lurks an exciting new arena for service management-virtualization and cloud computing.

The increase in virtualization has created a new challenge for service man-agement, which incorporates additional elements and processes behind traditional monitoring and management frameworks. Providing end-to-end management over this complex and heterogeneous environment becomes daunting.

Providing the visibility into virtual and cloud services will aid in real-time and trend measurements. This helps to achieve the ROI for cloud investments and strategy, holding IT accountable for the external service level agreements with the cloud and service providers that make up the extended IT organization.

Steps to Managing Your Virtual Environment

Vendors have developed a wide variety of monitoring and management tools to demonstrate how physical and virtual components are performing, so necessary course corrections can be made at the technology level. Unfortunately, all of these tools have separate purposes, data stores, formats, and interfaces. Data center managers need a complete view to analyze the data, for real-time decision making.

The Service Performance Dashboard

In complex, dynamic environments, it is imperative to consolidate and organize data from all physical and virtual systems. Whether you lease a virtual infrastructure from a service provider, or run your own virtual data center, tools exist that offer instant access to data. These tools integrate the overlay technology to connect the data in intelligent service models of a physical, virtual and cloud infrastructure.

For example, a service performance dashboard-a comprehensive dashboard of technology and business transactions-is designed to look at individual services based on certain set parameters. Signs come from the hardware, software, and business transactions against the service-level thresholds for availability and performance. Isolating each service on the dashboard provides a color-coded, graphical view of its end-to-end performance. Drilling down further exposes the root cause via the core data metrics, averting incidents and speeding restoration when an incident does occur, and zooming out provides a view of the virtual and physical technology supporting the troubled service, as well as the downstream services being affected.

Create Threshold or Operational Level Agreement

Building thresholds and algorithms into the model is an effective way to intelligently alert and manage service performance proactively. These thresholds propagate up the service model to determine the overall state of the service at any given time. They also incorporate business calendars to automate the prioritization process, as events occur within the infrastructure and/or business transactions.

Due to the relationships in the model, algorithms can be used to define rules that go far beyond multiple, separate monitoring systems. The metrics of various sources can then be weighted in setting overall service state and priority-meaning performance and/or business transaction performance might weigh higher than availability. In addition to averting service-impacting events, speed to restoration is quicker, prioritization is set in accordance to business objectives and the results feed into trend analysis for continuous service improvement.

Automating Governance

As the organization matures, the process of maintaining the model must be automated. This frees resources for strategic service analysis, instead of pure service monitoring. The operating model must be synchronized with the last approved model to ensure governance of approved change and model accuracy. This same synchronization also governs standard service and technology configurations, reducing support and management costs.

Another area of high-cost manual effort is the generation of service-level reporting.  The service performance dashboard must be a real-time, service-level management view of the service, in order to see and acknowledge impact, assess the situation and take action. Service-level reporting then should incorporate the value saved in averting these near-misses to create continuous service improvement.


In an environment where the physical and virtual collide, service managers can establish a unified service performance dashboard that isolates service performance, provides proactive notification and root-cause analysis, and evaluates its relationship to business objectives. Viewing technology through this service performance dashboard provides service managers and the business with an effective framework in managing today's new technology platforms.