The optimal performance of the network is what drives the most value in the corporate environment. Too often, we can’t complete business processes without the right tools in place, which are generally managed by the network. Fortunately, network management can go a long way in ensuring the readiness of the network and the subsequent performance.
Just how this performance is measured, however, is critical to the overall effectiveness of the network. A recent WebNMS blog examined the importance of tracking network KPIs in total network management. As stated in the blog, deriving actionable business intelligence from the network is critical to the improvement of its profitability and the meetings of established Service Level Agreements (SLAs).
To manage and optimize a network, it’s critical that the right Key Performance Indicators (KPIs) are defined, measured and quickly analyzed. These KPIs serve as visual cues that point to network performance and enable the administrator to make better and more informed decisions faster, so as to have the least amount of impact on the network and its users. This requires a focus on the holistic and forward-looking metrics that represent the complete network, a region or even a customer segment.
When KPIs are clearly defined, they help the administrator detect performance degradation and then fix problems before they can impact the customer SLA. In working with network management framework capabilities from WebNMS, administrators can build KPI scorecards for the network by way of a single console. For KPIs already being tracked, there is no longer a need to manually collate and analyze large complex data sets through multiple spreadsheets.
Now, administrators can use these KPIs to better see and understand the available data so it is better aligned with the goals set by the organization. Gaining such insight can help improve the speed and quality of the data-driven decisions. KPIs not only show how the network is performing – it also allows for the concentration of core responsibilities.
Understanding the importance of using KPIs is only the first step – administrators then need to decide what KPIs to use for network performance monitoring. KPIs can be established for the availability of the device, the interface, link, bandwidth and link performance.
For the device, it must be agentless and monitored with SNMP to verify that the CPU on the device can respond. Whether or not the device can be used is an indication that an outage may be occurring that will impact users. The same is true for the interface on each of these elements. For link availability, the source still needs to be agentless as most network devices do not actively report. The method preferred by vendors tends to like appliances that duplicate the functionality of Cisco (News - Alert) and Juniper.
Agentless is also required for bandwidth utilization and link performance. For utilization, people tend to use SNMP MIB2, as it’s commonly supported, and knowing the quantity of traffic over a link is considered vital information. For performance, the majority of administrators tend to use Cisco or Juniper to measure the quality of the link’s health.
Overall, network performance is still a priority for the diligent administrator focused on supporting users and ensuring optimal uptime. KPIs play an important role when identified and tracked correctly.
Edited by Blaise McNamee