Uptime is critical in service provider networks. With increasing competition, operators must keep their customers happy by delivering best in class services with minimal downtime. To do this, there are three network performance capabilities that every service provider must have.
Routing runs the Internet. Therefore, ensuring the performance of routing protocols, especially BGP, tops every service provider’s priority list. A common issue that can affect BGP routing is route leaks or route hijacks. Despite the advances in technology, we have seen numerous incidents over time, including the most recent major BGP leak by Google that resulted in outages in many countries.
The solution is to monitor the performance of BGP routing and be alerted when anomalies or deviations are detected. For example, alerts about potential BGP security risks, such as prefix changes or path changes, helps service providers proactively troubleshoot an issue before it causes in any major downtime. Monitoring routing performance also helps providers make better and more informed decisions to optimize peering.
Despite routing monitoring, a routing issue can bring down your network. This calls for immediate response – finding the root cause of the routing issue and resolving it. But routing issues are hard to troubleshoot. Some routing issues resolve themselves before a trouble ticket is raised, or there can be intermittent routing issues that are difficult to catch. In either case, the network team should be able to go back in time and see the path taken by traffic when the issue occurred. This is difficult, because most providers lack the tools to obtain historical data.
By monitoring and recording routing events and paths, a network engineer gains the ability to look at the actual routing path history with DVR-like playback. This helps to see how the network behaved and quickly identify the root cause of an issue, including those that occurred in the past and intermittent routing issues. This capability helps reduce MTTR.
Network configuration changes, either through human error or software bugs, are one of the top reasons for network downtime. Network engineers therefore need the ability to visualize the impact of a configuration change on the network before it is implemented. They also require the ability to analyze the result of a change after it has been made. This helps assure there are no surprises that can cause a downtime.
Using network planning, service providers can visualize the impact of a network change, be it adding new customers, redirecting traffic over back-up links or providing new services to existing customers. Network change management gives a before and after view of the network, helping determine if the planned network change has also caused unplanned network changes.
The Packet Design Explorer Suite delivers all three of these capabilities and more, helping improve overall network performance.