Capacity management can be almost anything your organization chooses to make it. Among practicing capacity managers, certain concepts and vocabulary are used to enable sharing of best practices. These first four chapters bring you into the dialog by sharing the concepts and vocabulary of capacity management
As an information technology (IT) manager, you must balance many forces to accomplish your mission. You need to hire and retain highly skilled employees without going over your personnel budget. You must be sure your security precautions protect against a wide variety of attacks without restricting legitimate users from accomplishing useful work. You need to balance the risk of new technologies against the features and functions your user community demands. Making trade-offs is the daily routine of every IT manager.
This book is fundamentally about another trade-off that every IT manager must face: how to deploy enough hardware and software to run the business without deploying so much that you are incurring unnecessary cost. The discipline of understanding and achieving balance between “too little” and “too much” is called capacity management, and whether you have a formal process with automated tools or you simply act on instinct, every IT manager is by definition a capacity manager.
Although capacity management has been an organized part of mainframe operations for many years, only recently have people been actively trying to manage capacity in the distributed server and workstation arenas. Many of the techniques learned in the mainframe data center still apply, but some important new concepts are required as well. In recent years, the community of people thinking critically about capacity management has begun to share best practices with one another, and those best practices have found their way into the set of all IT operational best practices, known as the IT Infrastructure Library or ITIL. The goal of this book is provide you with practical guidance in implementing those best practices for your organization.
This chapter provided an introduction to the world of ITIL capacity management. It began with a brief introduction to ITIL and explained that every IT service is part of the five-step life cycle:
strategy, design, transition, operation, and continuous improvement. The ITIL library provides a separate volume for each phase of this life cycle. Although capacity management is important throughout the life of every service, the process description itself is found in the ITIL volume on service design, reflecting the idea that capacity considerations should be designed into every service.
In the second section you learned why your organization should invest the money to build and mature your capacity management program. More than any other IT process, capacity management helps manage the costs of your IT infrastructure. Unless you already have very effective capacity management practices, you will get a great return on any investment you make in this process area. You saw that main goals of capacity management are to avoid overspending on resources and to avoid running out of resources when your critical business processes need them.
This theme of balance between spending and IT shortages is a continuing theme.
You began to explore the place of capacity management among your other IT processes.
One of the central tenets of ITIL is integration between process areas. Capacity management is highly dependent on configuration management and is a key enabler of service continuity management and release management. This topic of interplay between processes is so important that a whole chapter is devoted to it later in the book.
You looked at the business case for capacity management in some detail. You learned that cost avoidance is the biggest part of that case, because most organizations have already purchased enough hardware to last them for the next year or so if they will only manage their capacity more carefully. Consolidation and virtualization are strategies that enable you to free up space for future projects without spending more on hardware, and in some cases software. Good capacity management can also improve the ability of your organization to respond to new business conditions, and provide a great foundation for better management of your IT finances.
Finally, you took a quick tour of some of the risks associated with capacity management programs. Getting good projections from your business units on how much capacity they need can be problematic because many times the business is not accustomed to describing their needs in capacity management terms. The wide variety of components in the typical IT shop can lead to inconsistent measurements and metrics, making it hard to effectively manage the capacity of IT services that use many different components. Finally, the tools and skills needed to manage capacity effectively are not always available. Hopefully this book can provide a good start on the skills challenge, and an entire chapter is devoted to finding the right tools to help you. Being aware of these risks will help you to avoid them or at least mitigate their impact as you move forward.
Chapter 2 defines the terms and explains the concepts involved in managing IT capacity.
As with any new discipline, there is a certain vocabulary you’ll need to acquire. Fortunately, it is neither complex nor obscure. By the end of Chapter 2, you’ll be able to speak fluent capacity management!
One great way to think about IT capacity is in geographic terms. This chapter describes capacity as a series of pools and streams that can each be managed individually. Taken together, all the pools and streams in your organization describe the total of your IT capacity. Good management of all the pools and streams results in all the benefits of capacity management described in Chapter 1.
This chapter has described some of the basic terms involved in capacity management. You were first introduced to the idea of capacity pools, which enable the concept that ITIL calls component capacity management. You learned that capacity pools are groups of IT resources that are designed and managed in the same way. The boundaries of capacity pools can be defined by geography or organizational politics, but you learned that the broader the pool, the more effective you will be in managing it.
Capacity pools can be defined at a very granular level, but it is normally better to define your pools at a higher level to avoid the overhead involved in managing thousands of changes to the pools. You learned that managing capacity pools is a lot like managing the inventory for a manufacturing organization—you want to have resources available just in time and minimize the number of resources that are idle at any point.
The second section of the chapter introduced the concept of capacity streams. Whereas pools are groups of like resources, streams are groups of heterogeneous resources that work together to provide a common IT service. Capacity streams form the basis of what ITIL calls service capacity management. By looking at these streams, you can begin to understand how much capacity you need to provide an end-to-end service to your IT consumers.
You saw that capacity streams frequently intersect with one another because similar infrastructure resources are often shared between different IT services. When these streams form a confluence, capacity management should be especially careful to clearly identify the implications of shared capacity. Managing capacity streams, or managing service capacity as ITIL calls it, is more complex than managing simple pools of resources, but it is much more valuable. The difficult business questions are normally resolved by careful management of capacity streams rather than by management of capacity pools.
Because virtualization is so prevalent in today’s IT environments, this chapter included a section on how virtualization affects capacity management. You learned that virtualization creates more pools of resources to manage, but that modern hypervisor programs manage those resources effectively in most cases. The values that virtualization brings to an environment greatly outweigh the overhead costs of managing the virtualization engine.
In the next chapter, you expand on your growing knowledge of capacity management by looking at how organizations predict how much capacity they will need in the future. Accurately predicting the future needs of your organization will make you a much more effective capacity manager and allow you to take a much more active role in helping your organization reduce the cost of IT
Chapter 2 looked at how to measure the capacity of existing components and services. This chapter approaches ITIL capacity management from the opposite direction by examining the ways to predict how much capacity will be needed. Every dynamic organization goes through periods of greater and lesser need for IT services, and these cycles determine the demand for IT capacity.
Predicting the future needs of the organization will never be an exact science. In this chapter, you look at several methods for formulating an estimate, and hopefully some combination of these will be effective for your organization.
In this chapter you examined three separate techniques to forecast the amount of IT resources your organization will need in the future. The first technique simply relies on observing how many resources you’ve used in the past and assuming that your rate of consumption will stay rather steady. This can be surprisingly effective and generally requires very little extra effort on the part of the capacity management team. You learned that you can track component utilization, IT service utilization, and even business utilization using this simple trend-line approach.
A second way to predict future utilization is to look at the activities that your IT organization is planning for the future. These could be very large and visible projects such as implementation of a major new business application. On the other hand, you can also predict future utilization by looking at the day-to-day projects involved in upgrading operating systems, refreshing hardware, and maintaining your environment. Predictions based on planned IT activities have the advantage of being based in real anticipated work rather than an assumption that you will continue to do what you’ve always done.
Finally, you looked at business events as a way to predict future IT needs. While not nearly as specific as planned IT activities, it is quite often possible to predict growth in IT resource utilization by looking at the activities your whole organization will undertake.
You learned that the most accurate forecast will come from blending the various techniques.
Using everything you know about what you’ve done before plus anything you can learn about IT and business projects, you should be able to assemble a very complete and accurate projection of your IT capacity needs. After you have this accurate forecast, there are a number of ways to use it to manage your upcoming IT costs.
The next chapter details how capacity grows within an organization and some of the ways you can manage this growth as the capacity planner.
Thus far you have learned about capacity management in general (Chapter 1), the terminology used in managing capacity (Chapter 2), and the way to predict capacity needs (Chapter 3). In this chapter we turn to a deeper understanding of how capacity is consumed.
It might seem obvious that as the organization wants to do more, it requires more IT capacity. But that simple understanding isn’t nearly enough to be an effective manager of capacity. We must think more deeply about the dimensions of growth to really understand how to manage that growth in an effective way. This chapter guides you in that thinking.
This chapter takes the optimistic view that your organization is growing, and describes how to measure and manage the corresponding growth of your IT environment. In many cases your IT environment may be growing even if the overall organization is not. Tracking and managing IT growth is an important role for the capacity manager.
You first learned about base growth—the natural growth that happens as your IT systems are used and as they evolve to newer versions. You learned that the capacity manager is responsible for setting the upper limit for system growth because in many cases the computer systems and software don’t do that for themselves. You also learned that there is a gap between the way that tools report this growth and the way it should be displayed for management consumption. It is up to the capacity manager to bridge that gap.
You next looked at how business growth can impact IT capacity growth. You saw that the business can be expanding in one of two ways. It can provide new goods or services to its customers, which will generally grow revenue and probably expenses as well. When this growth occurs, there is almost always a resulting growth in IT utilization, which may involve growing the infrastructure. The other way that business can grow is by implementing internal projects to cut costs, enhance security, optimize compliance with industry regulations, or make the business more sound in other ways. These projects may or may not increase the demand on IT, and consolidation projects often reduce the IT demand. The savvy capacity manager needs to understand how the business is growing and be ready to respond with the best possible information to help the business make decisions.
The first four chapters have introduced you to the key concepts of capacity management.
We’ve set the context of capacity management against the ITIL framework, and you’ve learned how capacity management must interact with other ITIL process disciplines to be successful. You looked at capacity pools and capacity streams as a metaphor for what ITIL calls “component” capacity management and “service” capacity management. You’ve learned how to make predictions about future IT demand, and in this chapter you’ve begun to think about the ways that growth impacts your capacity management program. Now that you have a firm foundation, you can explore the best practices in capacity management as defined by the ITIL process.
In the next chapter you look at the first of those best practices—the capacity management information system (CMIS). You’ll learn why you need a CMIS and how to manage it. You will explore the CMIS in a way that doesn’t depend on any single tool or implementation approach, and then we circle back around to the best tools in Chapter 10, “Choose Capacity Management Tools.”
In Part I, you learned the general concepts about managing IT capacity. In this second part of the book, you learn the specific details needed to begin actively managing capacity for your organization. The chapters in this part delve into the details that are mentioned as best practices in the ITIL documents. By pulling from these best practices, you should have a solid base to implement your very own capacity management program.
The capacity management information system is a vital tool for understanding how much IT capacity is available to your organization and where you may need more. Inexplicably, many organizations do not yet have such a system implemented. In this chapter you learned not only what a CMIS can be, but also how to implement one.
At the beginning of the chapter, you learned that the purpose of the CMIS is to store all capacity-related data in one repository so that it can be integrated with other service management disciplines and so that you can produce capacity-related reports from an authoritative source.
Next you learned that the CMIS stores data on the utilization and capacity of your IT components and services, but that it also stores your capacity plans.
The middle part of the chapter described how to implement a capacity management information system using a very typical IT project development life cycle. You learned that the CMIS must be treated as a high-priority IT project and not just as staff work assigned to the operations team to handle.
This chapter concluded with some of the more obvious uses for the CMIS. Over time, you will find more uses, but initially you will use the CMIS to forecast capacity needs, make capacity decisions, and integrate capacity management with your other ITIL processes.
It is almost impossible to discuss the capacity management information system without touching on capacity plans. This chapter introduced the basic idea, but the next chapter dives into the details of what makes an effective capacity plan and how you can build capacity plans for your organization.
Capacity plans provide the capacity management team with the opportunity to analyze the mountain of data they work with every day and make solid recommendations to the IT management team. This chapter has described capacity plans and provided some insight into how your organization can create them.
We started with an overview of what makes up a good capacity plan. You learned that, in addition to specific data elements such as utilization and capacity limits, a solid plan should also include predictions and analysis. You should create a separate plan for each major IT service and for each significant IT component, with each plan providing an in-depth approach for what to do when capacity is too high or too low.
We considered the way to format your capacity plans. You learned that although data can be in the form of a table or structured text, your analysis and recommendations will almost always be free-form text within a document of some kind. You should provide an overview for casual readers, but also provide technical depth to justify the recommendations you make within the plan.
We considered how to manage capacity plans and saw that they follow a life cycle that can be tracked through effective document management practices. Capacity plans should be reviewed regularly and updated when changes are needed. Minor changes can be bundled together into a regular release cycle for the document, and major changes should be planned and implemented as part of the major IT change that is causing you to rethink the capacity plan.
As indicated in Chapter 5, capacity plans should be stored in the CMIS. If your CMIS tool allows attachments, you should store the actual document there. If not, you should at least create some kind of index within the CMIS so that users of that repository can easily locate and view your capacity plans.
Now that we’ve addressed the two most misunderstood ITIL capacity management terms (CMIS and capacity plan), the next chapter returns to fundamentals. We explore how to define and staff the roles you need to create an effective capacity management team.
Unfortunately, no one can tell you how to build the perfect IT organization. There are variations and exceptions to every rule. What I’ve tried to do in this chapter, however, is at least share the general rules with you so that you can begin to create the perfect capacity management organization for yourself. In this chapter you’ve learned the roles that other organizations have found useful and you’ve found some guidelines for staffing those roles.
You learned that the implementation team is led by the capacity process owner, who should be your most experienced capacity management expert. In addition, you need someone to design your capacity management information system, someone to design and document your process and procedures, and a project manager to keep the team focused in the right direction. This implementation team should be designed and staffed specifically to build the strongest possible foundation for your capacity management service.
But a firm foundation won’t last without maintenance. In the second half of the chapter, we explored the roles you need for your operational capacity management team. The leadership stays with the capacity process owner, who is now called the capacity service owner. This strong leader builds and often manages the team of capacity analysts who do the day-to-day work of understanding the capacity of your IT components and services. As the name implies, the analysts do more than generate data—they explore that data and make recommendations to the rest of IT. You also learned that a strong operational team includes a data manager who can manage the intricacies of the CMIS and a capacity planner who can summarize the recommendations of the various capacity analysts and bring them to a wider business audience.
I finished the chapter by giving you a few insights I’ve gained over the years about staffing, skill development, and maturity. These aren’t insights from a human resources professional, but from someone who has been in the trenches with the capacity management team and understands how the team can succeed if all the parts are in place.
In the next chapter we continue our survey of the best practices in capacity management by looking more deeply at the capacity management process and how to implement it. You’ll find that the process is what will bind your team together and keep them moving as a unit when difficulties arise.
This chapter described the steps involved in documenting and implementing a capacity management process. Starting with the high-level process flow adopted from ITIL or another process framework, you learned which attributes are important in a high-level process.
We next explored policies, which are aids in making decisions while following a process. I described several categories of policies and provided some guidance on how to define your policies within each of these categories.
Next you learned about procedures, which are more detailed steps that you document so that your team can follow the high-level process consistently. You learned about procedures for capacity reviews, procedures for improving capacity, and procedures for capacity planning, all fundamental steps in the ITIL capacity management process.
The chapter ended with instructions to help you manage all your process work. Processes, policies, and procedures need to evolve as your capability matures, and you need to have a plan to manage that change as it occurs.
In the next chapter we continue our survey of best practices by considering the relationship between capacity management and performance management. In thinking about your high-level process, you need to determine whether performance management is simply a subset of capacity management for you or whether it has enough scope to warrant a high-level process of its own.
The next chapter helps you to make that decision.
Performance management is a very detailed and demanding topic. This chapter has presented a broad overview of the subject with lots of simplification to help you understand the general theory. In the first section we looked at a general definition of performance management, and you saw that performance is really a measure of system capacity at some point in time. The next section described techniques for measuring and improving performance, along with a general procedure for attacking any kind of performance issue. The next section examined performance management for systems that exhibit considerable variability in their busyness, such as you would expect from business systems whose users work on only a single shift. Finally, we explored the relationship between virtualization engines and performance management and decided that, in general, virtualization helps automate performance management.
The next chapter covers capacity and performance management tools. You learn about the various categories of tools, what characteristics are important in each category, and the likelihood of finding good tools in each category. Rather than telling you which tools to buy, the next chapter gives you the information you need to make the best selection for your organization.
This chapter considered four kinds of tools that can help manage capacity in your organization. It described component capacity tracking tools and IT service capacity tracking tools and contrasted these categories with one another. There are many different tools that track components but far fewer that combine those components as IT services and track the capacity of an entire service.
You next looked at performance management tools. These run the gamut from simple performance reporting tools to sophisticated tool sets that simulate many users running a mix of transactions to enable you to do sophisticated performance testing before putting a new service or application into production.
You briefly considered capacity management information system tools, only to learn that there aren’t any strong offerings in this space. The CMIS is a relatively new concept, and the software publishers haven’t yet figured out how to build a system that meets all the requirements specified by ITIL.
This chapter described how to use the trade study technique to determine which tools are most appropriate for your organization. By putting some numeric values behind the importance of your requirements and the degree to which various tools meet those requirements, you can quantify the degree to which each tool fits your needs.
Chapter 11, “Produce Capacity Reports,” focuses on reporting as the final piece of an operational capacity management program. You learn about the different kinds of reports that are typically used and gain some insight into which ones will be essential for your program.
There are countless other reports that could be created to track capacity and measure the effectiveness of your capacity management program. This chapter has introduced you to some of the reports that have been most helpful to the organizations I’ve been involved with. Based on what individual trouble areas you have, your set of reports might include all the ones I’ve mentioned or might include a completely different group.
The important point of this chapter is not which reports you end up creating, but why you create them. You learned that generating data is easy, but producing useful information takes more work. You should consider what decisions need to be made and create reports that help people to make those decisions. Don’t put data elements on reports simply because you have access to a database with those elements in it. Instead consider the consumers of those reports and give them the information they need but no more than they need. In many cases the quality of a report can be greatly improved by reducing the amount of data on the report.
We looked at a set of reports that will help you manage the capacity under your control. You can produce a report that shows components with very high or very low utilization to help identify potential projects. A second report highlights the component trends to show where the utilization is growing or shrinking the fastest, which allows you to focus on things that might otherwise sneak up on you. We also considered IT service trend reports and looked in some detail at how to create artificial ways to calculate the utilization of an IT service. This whole set of reports provides high-quality information about the capacity you are managing.
In the second half of this chapter, we considered some of the ways you can measure and improve the effectiveness of your capacity management program. We looked at ways to measure the accuracy of your trends, because improving the accuracy of trends will help you predict more closely when you need to take capacity actions. We highlighted a way to track capacity-related incidents so that you can focus on reducing those incidents over time. Finally, you learned how to evaluate the completeness of your capacity plans. If your plans are complete enough, you can reduce the effort needed to implement them or perhaps even automate their implementation all together. All these reports help you to see your maturity as a capacity planning organization.
This chapter completes Part II, “Best Practices in Capacity Management.” In this part you’ve learned all the best practices for managing capacity. You’ve learned about the capacity management information system and how it is the base for all your critical information about capacity. You explored capacity plans and learned how to create and manage them. This part also described how to staff the capacity management team and implement the process that your team will use. You discovered the interdependency between capacity and performance management, and I made a case for why performance management should be viewed as a subtopic of capacity management. Finally, you learned about desirable characteristics of capacity management tools and how to use some of those tools to generate capacity management reports.
The next chapter begins the final part of the book, Part III, “Common Issues in Capacity Management,” which highlights some of the issues you are likely to face as you manage capacity for your organization. The first issue, highlighted in Chapter 12, “Business Capacity Planning,” is how to handle business capacity management. Thus far I’ve been pretty quiet about business capacity management, but ITIL insists it is a vital component of a complete capacity management program. In the next chapter you learn how to address this component, and I give you some hints for how to adjust if your business doesn’t want to be an active partner in managing capacity.
In Part I, you learned the general concepts about managing IT capacity. In Part II, you looked in depth at how to implement a capacity management program and how to make it effective over the long term.
This third and final part of the book tackles several common issues and challenges that you are likely to deal with. The issues addressed in this part include business capacity management, models for procuring additional capacity, capacity management within a project plan, and integrating capacity management with other IT process areas.
These are certainly not the only issues you will face, but they are the most common questions and issues I’ve seen in working with many of IBM’s largest corporate clients
In this chapter we examined business capacity management in greater detail. You discovered why it is difficult, and you learned why it is valuable. Overall, ITIL recommends business capacity management as an integral part of your capacity management program, and I concur. The clients that have implemented business capacity management on top of a robust IT capacity management capability have found the reward to be well worth the effort.
We began by discussing the scope of business capacity planning, and you learned that it is different than IT capacity management but is essentially an extension of the same techniques. You learned how to define a formula to calculate the utilization of any business process based on the utilization of the IT services supporting it, and we looked at how to use those utilization numbers to predict future capacity needs for your business processes.
We looked at the cost-versus-benefit analysis for business capacity planning. Although there can certainly be challenges, in most cases the benefits are greater than the costs. Obviously, if this isn’t true for your organization, you should be content with simple IT component and service capacity management until some factors change to make the benefit more valuable or to make business capacity management less costly.
You learned how to integrate business capacity planning with IT capacity planning and the value of building specific capacity plans for business units. And finally we touched on what to do if your business isn’t quite ready for full-blown business capacity planning yet.
The next chapter tackles another topic that challenges some beginning capacity management programs. You learn the relationship between capacity management and procurement and learn some techniques to help you order capacity on time, but not too early. One of the major benefits of capacity management comes when you avoid ordering IT components that will sit around without being used. After you learn the techniques described in Chapter 13, “Smoothing the Order Cycle,” you will have a much better chance of implementing a “just in time” ordering process.
In this chapter we’ve examined the link between capacity management and IT spending. You learned that for each capacity pool you manage, there needs to be a buffer of unused space. When that buffer becomes too small, you have reached your reorder level. Reorder levels are different for each kind of capacity you manage, and managing them is an important part of the capacity management program.
We looked at the factors that affect reorder levels and considered how your process maturity, vendor packaging, virtualization standards, and even the economy cause you to change the reorder levels from time to time. Keeping track of these changes and predicting just the right time to reorder hardware and software can be the hallmark of a well-run capacity management program. Failing to track the changes results in orders being placed too early, wasting money, or being placed too late, causing project delays.
You next discovered that how much you order at a time is nearly as important as when you place orders. There must be a balance between placing large orders with high costs and placing small orders that don’t add sufficient capacity.
Finally, we briefly considered what might happen if instead of ordering you have declining utilization and need to retire some capacity. We thought about what this might mean to your reorder levels and how to track the need to retire assets, and as a bonus you learned a few ways in which you might use excess capacity creatively rather than simply getting rid of it.
In the next chapter we look at another issue that confuses many beginning capacity managers. We talk about how the capacity management team should interact with project teams and project managers. I describe how the capacity plan should be created within the project life cycle and when it should be updated.
Every IT organization experiences the tension between an operational team that is trying to keep the environment stable and one or more project teams that are trying to introduce beneficial changes to the environment. As a steady-state capacity manager, you should pursue the opportunity to work with as many project teams as you can because every project will either consume some of your existing capacity or introduce new capacity to the environment. In this chapter we looked at the many ways in which the capacity management team can and should interact with IT project teams.
We considered four key deliverables of every IT project and thought about how those deliverables can be influenced and used by the capacity management team. You learned how to influence project requirements, design test cases, and implement capacity testing to make the project team aware of the impact they have on capacity management.
We looked at the elements needed for a capacity plan and considered how that plan could be developed as an additional project deliverable. Understanding capacity trends and helping the architect choose a plan for future expansion not only enables the project team to build a stronger system but also helps the capacity management team build a solid capacity plan with a minimum of guesswork required.
In the final section of the chapter, we recognized that not every project has the benefit of working with the capacity management team, and we considered some strategies for dealing with projects that neglect capacity concerns. It is never too late to manage the capacity of an IT system, even if that system is already in production operations.
In the next chapter we conclude our look at the operational issues surrounding capacity management. We specifically consider how the capacity management program and process should interface with other ITIL-aligned processes. By focusing on the intersection points between the processes and tools, you can reach beyond the capacity management program to improve the operational effectiveness of all IT processes.
Hopefully, throughout this book you’ve learned how to deploy an outstanding capacity manage-ment program and mature that program by continuous improvement. The theme of this chapter is that you can bring even more benefit to your organization by implementing other processes within the ITIL Version 3 framework. In fact, the more processes you implement and the more tightly you integrate them, the more benefit your organization sees.
You learned about the integration between capacity management and six specific processes in the ITIL framework. There are many more processes, and each can and should integrate with capacity management. These integrations shouldn’t be done simply because I say so or because your process engineers want to appear clever. Instead, they should be done because they provide significant business value to your organization and because they make your IT organization more effective at meeting the needs of the business. Over time, you’ll find that each new process you implement and each additional step you take toward tighter integration pays huge benefits and typically funds the next process or integration. As your capacity management process matures, you’ll find more ways to integrate it with the rest of your service management processes, and each one will bring greater value.
