Introduction

      When I type the Mappa.Mundi web address into my browser the html and graphics are seamlessly downloaded and the page displayed within a couple of seconds, but how does this data actually get from the server in Silicon Valley to my PC in a basement office in London? How does it flow through the Internet to reach me? You can answer this with traceroute. This is a useful tool that allows you to lift the lid on the Internet and get a packets-eye view of the network. If you're like me, intrigued as to how the Internet works beyond your browser and the telephone jack in the wall, then traceroute can be a fun tool with which to explore and map cyberspace [1].

      Traceroutes map the path that data packets take between two points in the Internet, showing all the intermediate nodes traversed, along with an indication of the speed of travel. The first traceroute application was created by Van Jacobson, at Lawrence Berkeley National Lab in the USA back in 1988, and the original release is still available [2]. Nowadays traceroute often comes as part of the operating system. For example, with Windows it is a small utility called "tracert", which is used from the MS-DOS prompt by typing:

tracert < Internet address, e.g. www.yahoo.com >

      As an example, I used traceroute to map out the path through the Internet from my PC in London to Mappa.Mundi.net. The program works away for a few seconds as it dynamically explores the pathway the data takes, hopping one node at time. The end result is this rather cryptic looking output:

      If you’ve never played with traceroute before this can look pretty meaningless, but it is in fact a kind of one dimensional map of the how the data flows, with each node traversed on a separate line. The map gives you valuable information on the real-time routing out in the Internet between London and Redwood City, CA (where the Mappa.Mundi.net web server is situated).

Triangulating the Net with Web-Based Traceroutes Conventional traceroute utilities are limited in one important respect. The origin point of the exploration is fixed to the location of your PC running the trace. To overcome this limitation, you can use Web-based tracers which allow you to run a traceroute from different starting points. This is very useful as one can explore the structure of Internet topology from multiple locations, a kind of virtual triangulation. As an example, clicking here will run a Web tracer from Canberra, Australia to Mappa.Mundi. There are several hundred freely available Web traceroute servers all around the world and Thomas Kernen maintains a good list of them at traceroute.org. » traceroute.org

      I tend to think of the traceroute output as similar to a railway timetable which charts how trains travel through the network, which stations they stop at and when they arrive. The traceroute "timetable" above shows that data travelling from London to Mappa.Mundi had to pass through eighteen intermediate "stations" (network routers) to reach the end of the line. The columns in the output are as follows, first there is a node number, followed by three columns of times - such as 20ms 30ms <10ms - measured in milliseconds. These are three separate measurements of the time it took for the data packet to travel from the origin computer to that particular node and back again. This is called the round-trip time and gives an indication of the speed of each link. The last column gives the name of the node as a domain name and numeric IP address.

      To decode trace "timetables" you read it line by line. In our trace we begin at node one which is the departmental router for my office which is known by the somewhat cryptic domain name cisco-2.bart.ucl.ac.uk. The data packets move rapidly onwards to the next node which only has a numeric address, and is likely to be an anonymous but important link somewhere in a server room in the university. At node number three the data leaves UCL's internal network and joins the London metropolitan network which provides a fast backbone for universities and colleges in London. From there, it is onto the Janet backbone (the UK's academic and research network) at a gateway machine called south-east-gw.ja.net. Node five is the gateway router to the transatlantic link for the Janet network, which connects to ny-pop.janet, the point of presence (pop) in New York. The data packets have crossed the Atlantic at this point and consequently you can see a marked jump in the round-trip travel times from under ten milliseconds to seventy milliseconds. The traffic then flows into the Teleglobe backbone network in New York at nodes seven and eight, before joining AlterNet (part of UUNET's backbone empire) also in New York. Notice the strange, long domain names of these routers which are at the core of Internet. The names contain useful location information that can be decoded. The large backbone operators tend to use these styles of naming conventions for their nodes that can guide the traceroute explorer. Packets flow through AlterNet from New York across the USA to Santa Clara, CA (node twelve) and then to Palo Alto, CA (nodes thirteen and fourteen). Again, notice the large leap in travel time associated with the extra geographic distance being covered. Node fifteen is the gateway to MIBH (Men in Black Hats), the ISP for Invisible Worlds, Inc. and Mappa.Mundi. Node nineteen is the end of the line, the magazine's web server. So for the Mappa.Mundi homepage to get to my browser it takes nineteen hops, crossing four different networks. Quite a feat of routing and cooperation that happens unseen to the millions of average Web surfers. [article continues on page 2]

» Turn to Page 2, and add a little geography