In recent years, it has become more complicated to effectively manage technology. For the most part, improvements in individual hardware and software components have made it easier to very quickly put together large and complex systems. Yet, there are still major challenges with applications and platforms concerning integration, cost, complexity, and scalability. If companies are to achieve real success in meeting these challenges, they must address, or build upon, systems management competencies.
Generally speaking, systems management is very broad in scope and can cover the management of operations, storage, network, systems, configurations, events, capacity, and performance. Taken together, these disciplines form one of IT’s visible contributions to a successful business environment. Regardless of an organization’s size, it is important to undertake formal systems management activities. For those who currently subscribe to this philosophy, there are additional actions that can be followed, systems management capabilities.
Obviously, the extent to which systems management is formally addressed will depend not only on the size and complexity of the IT environment, but also on economic constraints and the business setting. Systems management is not just for large IT environments; companies of all sizes can reap its benefits.
This article discusses why it is important to pursue these disciplines and reviews key aspects needed to build systems management competencies. In addition, I will examine some cost-effective ways to build, improve, and maintain these competencies.
PERFORMANCE & CAPACITY
Capacity planning and performance management form one related area of systems management familiar to many. The performance management discipline, specifically, is one that must be continuously practiced—not invoked only on an as-needed basis. Activities should cover day-to-day performance and event monitoring along with regular and ad hoc reporting capabilities, noting trends and reacting to performance issues when they arise. Performance methodologies should be primarily used to proactively manage formally agreed upon levels of service.
Often, it is easier, but not necessarily cheaper, to add resources such as CPU, storage, and bandwidth to address a particular performance problem. As a result, not as much time or energy is spent tuning performance-related systems and applications parameters when things are running smoothly. Tuning activity can fall dormant when performance and levels of service are acceptable. The challenge today is to be able to deliver end-to-end service over a mixed technology environment in the face of varying business and customer demands.
Intertwined with performance management is capacity planning. This discipline is easier to justify when capacity constraints negatively affect delivered service. Two useful and important thrusts of capacity planning are sizing technology resource configurations for both new and old applications, and performance prediction of applications. Over the years, significant time and effort have been dedicated to these areas; so much that an identifiable branch of technology industry resulted.
Configuration upgrade costs and known (or unknown) limitations of hardware and software configurations have driven the requirement for effective capacity planning. Investing in and devoting staff and related tools to planning for future requirements and wringing out more capacity and better performance from existing technology configurations enable IT to deliver predictable service levels over a more stable and known cost structure. These outcomes foster a favorable business environment and satisfied customers.
In recent years, other factors have emphasized the importance of capacity planning because of their effect on technology acquisition and ongoing support costs. Software license fees and structure, as well as the sheer number of and heterogeneity of network and server components, have exponentially increased the cost and complexity of most IT environments.