Throughout Every Layer

As system administrators, architects, and industry professionals, it’s quite difficult to look up from the day-to-day tasks that define our profession. However, the view from 20,000 feet can be incredibly rewarding, yet sometimes scary. In a few weeks, I’m presenting at a local conference about the present-day concerns with BGP (Border Gateway Protocol) security. In preparing for the conference, I’ve come to realize how much we take “the backbone” and the core infrastructure that makes up worldwide routing and transport, for granted. In particular, the modern-day security issues with a protocol such as BGP illustrate how vulnerable the “core” can really be. Abstract the relatively-small slice of convergence-based traffic that occupies global network traffic, and suddenly reality hits. As an obvious advocate of the furthering of convergence and VoIP-based technologies and solutions, I think we should always keep a 20,000 foot view of our entire infrastructure, VoIP or not. Obviously we don’t have control of every POP and distribution node that exists in our VPN between San Francisco and Miami, but we should be prepared for large-scale routing outages. Designing a “convergence disaster recovery” plan of sorts seems increasingly appropriate. Asking your ISP, telco, and VoIP service provider about their escalation procedures in the case of a large, multi-carrier outage or attack, is not out of line. Understanding every layer of what makes up your infrastructure, physical or virtual, can only improve the solutions and systems we deploy and administer. So, in my opinion, it’s important to be a “sponge” in this industry. Any knowledge we can gain in topics that are related, in any way, to what we do, is increasingly important. Overall, it’s truly incredible about what really happens through every layer when a user in Los Angeles calls another extension across a VPN in New York… and only in a matter of milliseconds.