These graphs show the number of unique clients seen by each individual server and by all servers combined, accumulated over the course of the 26 hour collection period.
The table below summarizes the data as of the end of the collection period.
These graphs show the number of unique clients seen in each 10 minute interval by each server and by all servers combined.
These graphs show the number of new unique clients seen in each 10 minute interval by each server and by all servers combined; that is, clients that had not been seen in any previous interval.
These graphs show the fraction of clients seen by all servers that were not seen by individual servers.
These graphs show the number of queries seen by individual servers. The graph for all queries shown here prior to 2002-09-06 was incorrect: it counted the queries according to the QDCOUNT field of the DNS header, rather than the actual number of parsable queries in the message. Some clients byte-swap the 16-bit QDCOUNT field, so the value 1 is incorrectly written as 256. This resulted in a largely inflated value in the "all query types" graph. (Perhaps these messages should be counted as invalid instead of as the query type they contain, since the DNS server rejects these messages.)
These graphs show the CDF of the number of request messages sent by clients to each server. With a logarithmic x-axis, we can see that over half the clients sent 8 or fewer messages. One client sent about 18 million messages, for an average rate of 192 messages per second over the 26 hour sample period. (The ~10M message client reported here previously was actually the second busiest client.)
Here is the CCDF of number of messages sent, with log/log axes.
Intersections and unions of client sets of pairs of root servers, with union of all servers for comparison.
Query types seen at all monitored servers (e, i, k, m), in full view and zoomed in. "Unknown" contains all non-standard query types; "other" contains all query types that had low counts. The numbers in the legend give the total number of queries of that type seen in the 26 hour sample period.
-- Ken Keys