While you’d be hard pressed to find a contemporary networking pro who still refers to the Internet as the “Information Superhighway,” the metaphor isn’t totally off-base. The Internet as we know it, after all, is a massive dynamic network of pathways that connect a global collection of enterprise- or ISP-owned Autonomous Systems (AS).
Data travels over the Internet like cars on a highway. But just like drivers heading down the interstate, traffic traveling over the Internet won’t blindly make it from point A to point B without a route in mind.
Human drivers need an understanding of real-world conditions like weather, congestion, or road closures via their GPS, for instance, to arrive at their destination on schedule. Similarly, the Border Gateway Protocol (BGP) gives monitoring solutions greater context into the journey of traffic commuting over the Internet.
Not unlike how drivers will plug in criteria in their navigation console that indicates certain conditions for their trip — ie. to avoid toll roads — BGP announces routing and reachability information from servers on the edge of each AS to allow routers to dictate delivery paths. It does not, however, tell you about performance over that route — traffic, pile-ups, or the condition of the road.
BGP and network performance monitoring
Not unlike DNS, networking pros and laymen alike tend to take BGP for granted, assuming that administrators can “set and forget” rules or network policies for travel within and outside of their AS, and traffic will flow accordingly.
BGP routers live on the edge of each AS, maintaining routing tables to direct packets in transit, along with a routing information base (RIB) containing information from internal and external peers that updates as “travel” conditions along a given path change. From there, BGP data allows routers to decide on the best path based on current reachability, hops counts, and even an organization’s own preferences when multiple paths are available.
Where BGP data comes in handy is in giving IT visibility into all of layer 3 hops and Autonomous Systems that an app may traverse on it’s path to delivery. With more and more companies leveraging some component of direct Internet access (DIA), they readily understand the first network their apps will encounter (their local ISP). But they lack insight into peering relationships, route changes, and abnormal network conditions that could lead to network degradation beyond their firewall.
While teams may have basic, binary insight into the success of delivery of an app travels end-to-end across the network via their SD-WAN, for instance, they won’t necessarily be able to identify all of the handoffs between ISPs and other network owners. This would leave enterprise IT blind to the owner of a degraded hop, for instance, and unable to make appropriate reparations.
This is all yet another layer of visibility that teams need — both locally and globally — to suss out factors that contribute to performance degradation for their end users. Because the internet is really a “network of networks,” there isn’t one master ISP that can control or manage all of the traffic. BGP data sheds a light on the consortium of players who makeup the internet, delivering context teams need to identify the root cause of performance degradations. When combined with monitoring data BGP correlation can illuminate upstream providers that companies may not be aware of or network bottlenecks that are masked by peering relationships.
To learn more about what insights IT teams need to assure end-user performance, read our latest whitepaper, Beyond Your Network: How Digital Transformation Fueled By Cloud Has Changed Performance Monitoring.