Capacity Management

One of the responsibilities of the service provider is to ensure that the service is able to cope with the demands put upon it. If the design is not able to cope, the service will fail to meet the performance specified in the SLA. In addition, the service must be designed to meet not only the requirements for today; it must anticipate future requirements and be able to meet them, too. Overspecifying the design would ensure that it could meet future demands; however, this may result in wasted expenditure if the actual demand does not come close to the capabilities of the design.

ITIL states that capacity management is responsible for ensuring that the capacity of IT services and the IT infrastructure is able to meet agreed current and future capacity and performance needs in a cost-effective and timely manner. The capacity management process has therefore to understand the likely changes in capacity requirements and ensure that the design and ongoing management of the service meet this demand. Delivering sufficient capacity is a key warranty aspect of a service that needs to be delivered if the benefits of the service are to be realized.

Capacity management is considered throughout the lifecycle; as part of strategy, the likely capacity requirements for a new service are considered as part of the service evaluation to ensure that the service is meeting a real need. In design, the service is engineered to cope with that demand and to be flexible enough to be able to adjust to meet changing capacity requirements. Transition ensures that the service, when implemented, is delivering according to its specification. The operational phase of the lifecycle ensures that day-to-day adjustments that are necessary to meet changes in requirements are implemented. Finally, as part of continual service improvement, capacity-related issues are addressed, and adjustments are made to ensure the most cost-effective and reliable delivery of the service is achieved.

Delivering What Is Required, When It Is Required

The purpose of the capacity management process is to understand the current and future capacity needs of the service and to ensure that the service and its supporting services are able to deliver to this level. The actual capacity requirements will have been agreed upon as part of service level management; capacity management must not only meet these but also ensure that the future needs of the business, which may change over time, are also met. An essential objective is to deliver any increased capacity in time so that the business is not impacted.

The objectives of capacity management are met by the development of a detailed plan that states the current business requirement, the expected future requirement, and the actions that will be taken to meet these requirements. This plan should be reviewed and updated at regular intervals (at least annually) to ensure that changes in business requirements are considered. Similarly, any requests to change the current configuration will be considered by capacity management to ensure that they are in line with expectations or, if not, that the capacity plan is amended to suit the changed requirement. Those responsible for capacity management will review any issues that arise and help resolve any incidents or problems that are the result of insufficient capacity. This helps ensure that the service meets its objectives.

As part of the ongoing management of capacity and its continual improvement, any proactive measures that may improve performance at a reasonable cost are identified and acted upon. Advice and guidance on capacity and performance-related issues is provided, and assistance is given to service operations with performance and capacity-related incidents and problems.

What Should Capacity Management Include?

The capacity management process has responsibility for ensuring sufficient capacity at all times, including both planning for short-term fluctuations, such as that caused by seasonal variations, and ensuring that the required capacity is there for longer-term business expansion. Changes in demand may sometimes actually be reductions in that demand, and this is also within the scope of the process. Capacity management should ensure that as demand for the service falls, the capacity provided for that service is also reduced to ensure unnecessary expenditure is avoided.

The process includes all aspects of service provision and therefore may involve the technical, applications, and operations management functions. Other aspects of capacity, such as staff resources, are also considered.

Capacity Management in Action

A retail organization that was struggling to maintain its market position decided to expand its online and telephone-ordering service through a major marketing campaign. As part of this initiative, the telephone-ordering service hours were planned to be extended to 24 hours, 7 days a week. The business was considering what this would mean in terms of increased call-center staff, warehouse staff, and stock levels. The IT director was tasked with ensuring that the IT services would support this business initiative.

The IT director called together his managers involved in the capacity management process. Those in the technical management function had to ensure that the infrastructure would be able to cope with the expected increased demand. This included ensuring the telecoms infrastructure capacity required for the extra call-center staff and the voice traffic that the staff would generate in addition to the increase in data traffic. The website’s capacity to handle increased traffic and the ability of the applications to handle a high volume of orders, credit-card processing, and so on, were investigated by the technical and applications management functions. The technical solutions that were recommended as a result meant more equipment would be purchased. The operations management function investigated the impact on operational processes, such as increased time needed to carry out backups and the impact of 24/7 operations on planned maintenance. Included was the impact on the UPS, air conditioning, and so on, of the extra equipment. Finally, the service desk manager calculated what increase in staff would be required to move to a 24/7 support operation and an increased user population during peak hours. This was calculated as requiring two new service desk analysts, and the building services department was asked to provide the extra office space for the new staff.

---

As this example illustrates, an increase in capacity requirements may have repercussions across the infrastructure and on the IT staff resources required to manage it. Although staffing is a line management responsibility, the calculation of resource requirements in this area is also part of the overall capacity management process.

Capacity management also involves monitoring “patterns of business activity” to understand how well the infrastructure is meeting the demands upon it and making adjustments as required to ensure that the demand is met. Proactive improvements to capacity may also be implemented, where justified, and any incidents caused by capacity issues need to be investigated.

Capacity management may also recommend demand management techniques to smooth out excessive peaks in demand.

Managing Capacity by Managing Demand

Like many countries, a certain European country requires that taxpayers complete a tax return online by a particular date. This is enforced by an automatic fine for late submissions. Inevitably, most taxpayers leave this until the last minute, with a huge demand for the system being generated in the last few days before the deadline. This demand profile means that the systems are overengineered to cope with a short-term spike in demand, and yet there is still a real risk of insufficient capacity in the last few days.

Providing even more capacity is not a good option to deal with this situation, because the design is overspecified for the level of demand across most of the year. Instead, an element of demand management is introduced. In addition to the 100-euro fine for late submission, an incentive is introduced; for every day before the deadline that the tax return is submitted, there is a one-euro reduction on the tax bill. This motivates people to submit well before the deadline, reducing the maximum capacity required and avoiding any capacity-related incidents.

---

The Capacity Management Subprocesses

To draw up the capacity plan, you must understand the future requirements of the business. Business capacity management is therefore a subprocess of the overall capacity management process. Its aim is to calculate what the business plans and forecast mean for the infrastructure. Business plans to expand (perhaps by buying smaller competitors) or launch a new service, offshore a call center, or outsource a function will all have an impact on capacity requirements. Capacity management has to understand this impact on the infrastructure and plan to meet the changed demand.

In addition to understanding overall business plans, capacity management must understand how the use of individual live services may vary over time. Service capacity management is the second of the capacity management subprocesses. The service level requirements for each of the live services must be understood, and monitoring needs to be implemented to check how well the service is performing. Thresholds can be set within the monitoring tools to enable any possible issues to be spotted and acted upon quickly. See Chapter 12, “The Other Service Operation Processes,” for more information on event management and the use of monitoring tools and thresholds. Capacity management will take action to handle any issues that arise and will also consider any proactive steps that could be taken to improve performance. This may require working with the specialist staff members who carry out component capacity management.

Overlap with Business Capacity Management

Service capacity management focuses on delivering the required and agreed capacity for each service. Monitoring performance will help you understand any changing use of the service, which overlaps with and feeds into the business capacity subprocess. Even if overall capacity requirements are increasing, there may be a reduction in the demand for a particular service. For example, a car insurance company that was launched some years ago to sell insurance directly to the public, through large call centers, may find that although business is increasing, the use of the call center (and the IT services used by the call-center agents) is decreasing, being replaced by online sales. Monitoring the increase or decrease in the use of a service will help business capacity management to understand the likely requirements of the service in 12 to 18 months, which can be fed into the overall capacity plan.

---

The final subprocess of capacity management is component capacity management. This is the most technical aspect of capacity management and is likely to be carried out by the technical management staff, with day-to-day monitoring being the responsibility of the technical and operations management functions. It requires a detailed understanding of all the components that make up the end-to-end service and their individual capabilities. The current utilization of each of these components must be monitored, because insufficient capacity in a single component could cause a bottleneck and impact the whole service. This subprocess will use event management to track when thresholds are breached so that preemptive action can be taken to overcome the issue, before service is affected.

Figure 6.9 shows the three capacity management subprocesses and how each is involved in one or more of the capacity management processes steps of reviewing and improving current capacity, identifying changes to capacity requirements, and providing new capacity. The figure also shows the capacity plan, which we will discuss next.

FIGURE 6.9 Capacity management overview with subprocesses

Based on Cabinet Office ITIL® material. Reproduced under license from the Cabinet Office.

Managing capacity requires the IT service provider to respond to changing circumstances because the expected capacity demands may prove to be an under or overestimation. This requires a number of key activities to be performed iteratively as part of the service operation stage of the lifecycle. Figure 6.10 shows this ongoing cycle. The first step involves monitoring capacity against thresholds and then analyzing the data this provides. Any necessary “tuning” adjustments, such as adding or removing resources in order to ensure that sufficient capacity is always available, are identified. These are then implemented, and the monitoring stage begins again. All of the information gathered from these activities is stored in the capacity management information system (CMIS). The CMIS forms part of the overall service knowledge management system (SKMS).

FIGURE 6.10 Iterative activities in capacity management

Based on Cabinet Office ITIL® material. Reproduced under license from the Cabinet Office.

Planning for the Future: The Capacity Plan and CMIS

Capacity management works with the business to understand its current and future business needs and investigates future technical developments that may be able to help provide capacity more cheaply; this information will be used to draw up the capacity plan.

The capacity plan is an important output from the capacity management process. It captures the current and future requirements and proposes how these should be met. Drawing up the plan requires a close relationship with the strategy phase of the lifecycle; the strategy needs to be based on a firm understanding of the current and future capabilities of the infrastructure, and the planners need input from the strategy discussions to understand what the future requirements will be.

The plan covers 12 to 18 months ahead so that any planned expenditure can be included in the negotiation of the annual IT budget. Planning this far ahead does mean assumptions have to be made, so the plan should be reviewed and reissued as actual requirements become clear. Reviews should take place at least annually and, additionally, after any major business change (decision to expand or divest, outsource, offshore, or otherwise change the IT requirements). Each review of the plan should include a reforecasting of requirements for the 12 to 18 months following the review.

The plan is a resource for the business and all IT departments, which will both contribute to it and refer to it when planning services. It should contain an introductory section explaining the current infrastructure and its performance and any capacity issues currently being experienced. It should be clear what is in scope and what is out of scope so that the reader understands any omissions and the reasons for them. Any assumptions made should also be stated.

The plan should consider a number of possible scenarios and explain why these have been chosen. These scenarios should be based on the information that has been gathered as part of the evaluation of business, service, and component capacity requirements as being reasonable possible outcomes. For example, the business may be planning to increase the workforce by at least 10 percent in six months, based on expected increased sales. The scenarios used might include not only one to meet the 10 percent increase but also one for a 5 percent increase and one for 12 percent so that if the business is less or more successful than it expects, there will be a plan for providing the required capacity.

The plan should examine the current capacity demands of each service, how these are met, and the forecast for future capacity requirements, based on the scenarios. This may be at a detailed level, considering storage, bandwidth, and so on. Options for meeting the new requirements should then be described, together with costs, risks, advantages, and disadvantages of each. Finally, the plan should recommend a particular approach. It is essential that changes are assessed for impact on the capacity plan and on the performance of all services and resources.

The capacity plan is stored in the capacity management information system (CMIS), together with reports of current performance and forecasts of future requirements.

Effective capacity management is essential if the service is going to be able to meet current and future demands. All three areas must be considered—business, service, and component capacity—to ensure that the process fulfills this requirement. Poor capacity management will have a serious effect on the ability of the business to achieve its objectives, either through a failure to deliver the service required or by making the service unnecessarily expensive by overproviding capacity.