Guaranteeing Fast Service

In this Issue:

United Airlines' flight operations center has saved $46 million in operational costs related to fuel, staffing, and overtime by optimizing its ability to change course and shift resources in real time. The ability to proactively respond to changes in business is becoming paramount in industries whose very natures are fast moving and mission critical: Telecom companies and service providers are expected to deliver unwavering voice quality or data speeds. The defense industry needs real-time assessment and dissemination of intelligence, command, and control. The transportation industry has to retool and shift assets immediately when flights are delayed or flight paths change.

To quickly accommodate their enterprise customers' shifts in demand, suppliers that need to retool and reallocate resources are relying heavily on service-level agreements (SLAs) with real-time parameters. In enterprise applications, such as supply chain management and corporate performance management, SLAs must have a certain degree of flexibility if they are to keep up with step changes in intelligence and data technologies.

"Our customers want SLAs that dictate how we will evolve to support changing business conditions," says Joe Hogan, VP strategy alliances and marketing for managed services within Hewlett-Packard's Customer Solutions Group, which is focusing on creating flexible service levels. "The environment is never static; you dont want to renegotiate every time there is a step change in technology."

Hogan warns there must exist a balance between being flexible and being specific. "In ensuring that service levels support service capabilities, you can't just say 'I want 99-percent uptime,'" says Hogan. "Rather, you need to state 'I want tailored tutorials at service representatives' fingertips, so they can share the same environment in real-time with a caller having problems.'" It's then up to the supplier — the IT organization or manufacturer — to say it requires 99-percent uptime to deliver that business process.

As service-oriented architectures and emerging technologies such as radio frequency identification pick up momentum, cost structures will mature and prices will come down. IT organizations will have more choices in terms of creating technology events. "It will be flexible application layer and infrastructure layers that will enable IT to support mission-critical business processes," remarks Hogan.

When companies think of on-demand or adaptive computing, they think of IBM and HP "stack" solutions that resolve the hardware part of the equation. However, the demand for fluid environments, where applications and resources are shifted in real time, requires a focus that goes beyond just hardware and into the actual business. Real-time business will rely on the integration of capabilities from application, network, database, and server stacks — all of which work cohesively in on-demand environments.

The first layer of database technology must be performance-oriented and real-time in nature, as must be the servers and infrastructure. Sybase, for example, recently announced a real-time data services solution. By extending replication, data will be synchronized from different hardware across the enterprise.

"With more users, vendors, and partners demanding responses in milliseconds, high-performance data management and messaging technologies are growing in importance," says Noel Yuhanna, industry analyst, Forrester Research. Because data management components, integration of data, and real-time data management will be paramount, companies like TimesTen (vendor to United Airlines) and KX Systems are developing software that picks up the slack where traditional data repositories lack the high speed and scalability required for high-volume analytics and dissemination of information. Yuhanna believes a standard will ultimately evolve for data integration and data movement across enterprises. "Something analogous to the Enterprise Grid Alliance," he ruminates.

In the meantime, data integration and real-time data sharing will rely heavily on parameters defined in SLAs — parameters designed around business processes and goals.

Susana Schwartz is a New York-based freelance writer specializing in emerging technologies and their impact on IT infrastructure.