Service-Oriented IT Operations – A Framework for Improved Service Delivery
March 13, 2013

Vic Nyman
BlueStripe Software

Share this

"It's tough to be strategic when your pants are on fire."

This is actually a quote from Ron Kifer, VP of Global IT at Hewlett-Packard. His quote is a perfect way to describe the core challenge facing senior IT leaders.

Companies expect their senior IT leaders to be strategic, but their organizations are so busy with firefighting that they can't deliver on innovation.

No matter how brilliant the idea, it won't get implemented if the IT organization can't deliver the basics — applications that run the way end users expect them to. Between the shift to Agile development (greatly increasing the speed of the development cycle) and the move to make business applications the focus of so many companies' growth plans, the pace of change for IT Operations is outstripping their ability to execute. To help companies get their operations in shape to truly support business requirements, BlueStripe is advocating a service-oriented approach to application management.

What does “service-oriented” mean?

It means that IT needs to focus hard on the delivery of application service levels as experienced by end users. This seems obvious, but a surprising number of companies still manage their operations by measuring and managing resource levels or just managing the application code — creating a disconnect between what IT teams are expected to deliver and what they have the ability to manage. But when companies make the switch away from server monitoring to service-level management, the results can be dramatic.

The Service-Oriented approach to IT Operations is based on the best practices that emerged from our work with hundreds of customers, and involves three main concepts in how to manage applications.

1. begin with the end user in mind

First (to paraphrase Steven Covey), begin with the end user in mind. Instead of focusing on machine resource utilization metrics, monitor and manage transaction response times. In a distributed, virtualized environment, high CPU (or other similar resource metrics) are likely to be meaningless. By focusing first on transaction response time, IT Operations teams can put their efforts where they will have the most impact on the customer or end-user experience.

2. recognize that the entire application infrastructure can impact performance

Second, recognize that the entire application infrastructure can impact performance. The traditional, and outdated, application performance management (APM) approach is designed around identifying and fixing code-level issues in the application server. These tools have their place, but the Service-Oriented approach expands the view to the rest of the application infrastructure – the web tier, authentication, the database, and even third-party systems. Every component impacts transaction response times — make sure you look at all of them.

3. recognize that every infrastructure layer can impact component response times

Third, once the problem component is identified, recognize that every infrastructure layer can impact component response times. It isn't enough to look just at the application layer, itself. Every infrastructure dependency of that server is a potential culprit. Effective Ops teams must be able to see the target application, other applications running on the same server, the operating system, the virtualization layer, storage, networking, shared services like DNS, and even management tools that run on the server.

In practice, the application of these concepts requires the ability to track individual transactions across the infrastructure and the ability to measure response times between components. With the ability to isolate the bottleneck component, IT Operations teams can then pivot to drill into the server stack and determine the true root cause of the failure.

The Service-Oriented approach to IT Operations gives IT teams the ability to deliver proper service levels with greatly reduced effort. There will always be some amount of firefighting, but the Service-Oriented approach frees resources and builds credibility — so that IT can move from being a cost center to being reliable business partner.

Vic Nyman is Co-Founder and COO of BlueStripe Software.

Related Links:

A successful software veteran with over 20 years of systems management experience, Vic Nyman is a recognized expert in systems management and IT Operations. Before founding BlueStripe, Vic served as Chairman and CEO of Relicore, leading the company to a successful merger with Symantec in 2006. Prior to Relicore, Vic was the first business executive at Wily Technology, helping lead the company to market leadership. Vic has also held several executive positions with IBM in the Tivoli and Networking Software organizations. He has a BS in Electrical Engineering from the University of Illinois and an MBA from Duke University's Fuqua School of Business.
Share this