By Peter Christensen, Product Marketing Manager, APM, InfoVista
It is no surprise that traffic on service providers' networks is exploding. Recent analysis found that global voice and data traffic grew more than 200 percent in 2011 for the fourth year in a row. As a result of this unanticipated traffic growth, communications service providers (CSPs) are struggling to anticipate future network growth and ensure that prudent investments in growing capacity are made.
Currently, most capacity planning efforts are done manually and require extensive data gathering from multiple sources. Unfortunately, where these efforts typically fall short is the limited information available from simple network management protocol (SNMP) or from the device command line interface (CLIs). While this data is useful to understand utilization trends and short term statuses for individual interfaces, it does not provide a deep understanding of the site-to-site traffic matrix. This can result in an inadequate assessment of present and future capacity needs, frequently resulting in poor network performance and upset customers. To avoid this, CSPs need to take advantage of flow-enabled capacity planning.
Because it is already implemented in most backbone routers, the upfront investment for flow-enabled capacity planning is minimal. One alternative, network probes, is expensive and can be limited in their capability to handle large amounts of traffic in service provider backbone networks. Using flow data, it is possible to gather information about any traffic that traverses the network, including:
- How much traffic over a given service (voice, GPRS, corporate or terms of service) is transferred between two sites;
- Which protocols or services are using the most bandwidth;
- Whether VoIP traffic is impacting specific quality of service (QoS) or quality of experience (QoE) requirements;
- When mobile data traffic volumes peak;
- What the average traffic volume for the past week, month or year was;
- What percentage of total backbone traffic is P2P transit;
- How much traffic is there between two remote Gateway GPRS Support Nodes (GGSNs);
- How much traffic is there between a GGSN and the Internet; and
- How much traffic is generated between two customer and two corporate sites.
Armed with the information provided by a flow-based capacity planning solution, CSPs can quickly identify tactical issues and address them before customers are impacted. Modeling the impact of new services on available bandwidth before they are deployed, CSPs can decrease risk, increase revenue potential and ensure that problematic issues are addressed before rollout. All strategic investments in infrastructure can be supported by factual data rather thaneducated guesses.
To learn more about capacity planning, view our video, "Best Practices for Implementing Backbone Capacity Planning":