Modeling involves performance and capacity prediction through estimation, trend analysis, analytical modeling, simulation modeling and benchmarking. Modeling can be performed for all or any layer of the IT solution including the business, application and technology infrastructure.

Application sizing is a technique that predicts the service level requirements for response times, throughput, and batch elapsed times. It also predicts resource consumption and cost implications for new or changed applications. It predicts the effect on other interfacing applications. It is performed at the beginning of the solution life cycle and continues through the development, testing and implementation phases. Application sizing has a strong correlation with performance engineering.

Performance engineering is a technique that focuses on the assessment, establishment, and integration of performance planning processes and performance engineering methods within the development life cycle and implementation of prepackaged software. Performance engineering can aid in planning for effective use of existing resources, making informed equipment purchase decisions, and addressing potential performance risks and exposures more quickly. To improve strategic planning and reduce development costs, performance engineering methods and practices can be incorporated into the application development and business planning processes.

Understanding application design implications, system requirements, capabilities and costs early in the application development process improves project planning to help ensure success. Using these processes also helps your staff continually improve system performance, reduce costs, and increase productivity and user satisfaction.

Modeling and sizing are used to determine performance and capacity requirements. These requirements are met by the formulation and implementation of policies. Establishing and maintaining Performance and Capacity Management policies involves administration of pools of specific computing resources by managing policies for how resources are reserved, whether overbooking is allowed, how resources are monitored, and so forth. Resource-specific policies depend on the characteristics that are associated with particular resource types. For example, storage systems have different characteristics (space allocated, striping, access control) than networks (bandwidth allocation, packet loss rate). A policy framework provides a general, formalized way of controlling such customization and variability within a system through the use of policies.