Traceroute: Ping on steroids

Segment 6 of 7

   

At the end of the process, Traceroute generates a report that looks like this:

This example shows that it is commonplace for each router step to be repeated 3 times so that transit times can be averaged for better accuracy. It's also interesting to note that each - "hop" - tends to get a little faster and more efficient as the test proceeds. This is because each intervening router's high-speed - "cache" - memory remembers details of prior exchanges for a few seconds, and doesn't have to perform as many "housekeeping" or "setup" chores for the subsequent operations (so long as they occur before some other traffic crowds the records out of that cache memory).

This example shows that mit.edu is just 5 hops away from our PC. The first hop is always forwarded by our own LAN's NAT router, which tends to respond quickly.

The second hop is forwarded by our Boston-based Internet Service Provider. (Note that each hop is exercised 3 times, and that the first time our ISP sees our message, it needs 35 milliseconds to respond. After that, it is able to respond a bit faster, probably because some of the information it needs to construct and address the response is still in it's temporary, high-speed cache memory).

The Third hop is forwarded by an anonymous router at IP address 4.79.2.2, for which no DNS name is available.

The fourth hop comes from a router at IP address 18.168.0.25, which has been assigned DNS name - "W92-RTR-1-BACKBONE.MIT.EDU" - implying that it resides on the university's campus.

The IP address of the final hop is so similar as to imply association with MIT, and the DNS name is "WEB.MIT.EDU". That name is probably registered to the same computer as mit.edu in the Internet's Domain Name Servers.

From the information in this traceroute report, we can see that our computer enjoys a very good connection with the web server at mit.edu, and that the average round-trip time is a bit less than 49 milliseconds.

Next Segment  Prior Segment