This chapter is from the book
This chapter is from the book
This chapter is from the book
The Management Layer
For every layer described earlier, there is a discrete management function. Storage management consists of such low-level functions as storage device status monitoring to higher-level functions such as proactive capacity planning based on usage. Server and network management each entail similar low- and high-level functions, including server and network device configuration, performance threshold setting, threshold monitoring, and trend analysis for capacity planning. Application management, as described earlier, includes process monitoring, resource consumption monitoring, response time monitoring, and the proactive detection and resolution of potential problems.
With all of this layer-specific monitoring and management going on, one might wonder why a separate management layer exists on the figures contained in this chapter. It is a good question.
IT services, as discussed in the preceding pages, are enablers of mission-critical business processes with which they are increasingly intertwined. As such, IT services need to be held to the same performance standards as business processes themselves. They need to be managed for the quality of service that they provide to the organization, quantified and measured, subjected to guaranteed service level minimums, then monitored to ensure that service levels are met.
The management layer in this model embodies more than layer-specific management functions. A business manager is not particularly interested in the heat generation of a specific disk drive in a specific array located in a specific zone of a storage area network. Nor is the manager anxious to see a report on the number of Internet protocol (IP) packets dropped by a particular port of a network switch, or the CPU utilization of a specific server in the "server farm."
What does interest the business manager, ultimately, is how much improved productivity has been realized as a consequence of IT services. What percentage of customer requests are being handled within the first 24 hours of receipt? How many customer inquiries are being addressed via automation as opposed to requiring expensive "live" conversations with a customer support operator? How much have sales revenues increased as a result of the new e-commerce site? How has just-in-time manufacturing, augmented by automated supply chain management, reduced inventory costs and improved market share?
These questions, and myriad others, cannot be asked of infrastructure-level management resources. A management layer thus appears in the diagram to symbolize the tools and techniques emerging within IT services over the past several years that aim to distill platform and application management data into reports, analyses, and online representations that depict the impact—and hopefully the payoff—of the organization's IT investment.
At a minimum, capabilities should be provided to enable business management to view IT services on a business process by business process basis, so they can readily see whether the IT services supporting a specific process are doing the job. Performance measures, carefully defined, must be harnessed to provide a snapshot of the quality of service that is being provided at any given time. Historical data about performance needs to be wedded to service level agreements and to budgets to identify the compliance of IT services with both and to quantify cost differentials between planned and actual IT service performance.
IT services are not complete until the requirements of the management layer have been fulfilled, as shown in Figure 1–6.
){kind=link}
Figure 1–6 Management infrastructure requirements.