Closely associated with network performance management, capacity planning determines the network resources required to support business-critical applications and prevent performance problems. Capacity planning considers how the network baseline (bandwidth and usage, etc.) compares with network traffic trends in order to anticipate capacity expansions. As traffic on a network increases, consuming its capacity, network delays and jitter become more apparent, especially in real-time applications like VoIP and streaming services. Capacity planning is challenged with balancing the appropriate amount of resources that will ensure a quality user experience, even in peak times, while remaining cost effective.
Network Performance Management and Capacity Planning Best Practices
Because performance problems are usually grounded in capacity, capacity planning and network performance management best practices have significant overlap. The following best practices are shared between the two.
- Service Level Management — Network managers should create service level agreements between users and network organizations which include capacity and performance management. Understanding that service level agreements are living documents, they age and their terms become irrelevant or obsolete over time, a feedback mechanism must be included. By simply fully defining the service level objective metrics, sharing these metrics with users, and garnering their feedback, the loop is complete. Network managers can then set performance goals based on feedback or make recommendations for further capacity planning.
- Network and Application What-if Analysis — Before adding a new solution to the network, network managers should document a what-if analysis of the solution and its alternatives. The main objective is to document the risks, requirements, major questions, methodology, data sets and configuration files to a degree that it can be recreated by someone else. In the case of network what-if analysis, recreating the production network in a lab and modeling the scenario possibly on routing information borrowed from the production network. Or for application solutions, by using tools such as WAN delay simulators to simulated distance, these models should be able to be recreated within a lab using documentation. Through this analysis, network and application groups will have a better understanding of what could cause poor performance and prevent unsuccessful deployments.
- Baselining and Trending — The establishment of a baseline and the monitoring of network trends helps network managers anticipate capacity problems and plan on upgrades before they become issues. Baselining and trending is a proactive approach, and is complicated by the overwhelming amount of network information required for review in large networks. Furthermore, network management solutions provide information and graphic tools on capacity resource variables but are often used reactively to solve existing problems which defeats the purpose of baselining and trending. Someone must review and perform comparative analysis of resource utilization information on a weekly, bi-weekly, or monthly basis.
- Exception Management — Exception management is a methodology for anticipating and resolving capacity and performance issues. The system notifies network managers of capacity and performance threshold violations to be investigated and fixed. The near real-time notifications aspect of exception management allows problems to be seen and addressed before they disappear. The ability to anticipate impending performance deterioration is a hallmark of an ideal network management system.
- QoS Management — Quality of Service (QoS) management approaches traffic prioritization by classifying network traffic into classes. Uniquely this allows network managers to provide more consistent performance to specific application groups (traffic classes) by prioritizing network resources to those classes. Defining these classes requires a baseline understanding of network utilization, specific application requirements, and business application priorities that combined determine the appropriate bandwidth usage for each class.
Network Performance Management
Network performance management is the practice of monitoring a network and mitigating performance issues in order to maintain a required network performance level. Network monitoring plays a pivotal role in network performance management. It helps to identify performance bottlenecks by routinely reporting on the performance and quality service level of network devices such as routers.
Network Performance Baselining
Much like the balance sheet is a snapshot of an organization’s financials at one point in time, a network performance baseline is a picture of a network infrastructure’s traffic during normal working conditions. Network managers use baseline performance metrics as an image to compare to abnormal traffic conditions in order to anticipate network deterioration or failures. Network performance baselines are also important for determining when future capacity upgrades are necessary, or for complying with the capacity requirements of service-level agreements (SLA).