In the discussions on bufferbloat so far, an astounding number of papers have been referenced:
Read the slides starting with “If RED is not good enough, what is?”.
Netalyzr: Illuminating The Edge
As outlined in the IMC Netalyzr paper section 5.2, the structure you see is very useful to see what buffer sizes and provisioned bandwidths are common. The diagonal lines indicate the latency (in seconds!) caused by the buffering. Both wired and wireless Netalyzer data are mixed in the above plots. The structure shows common buffer sizes that are sometimes as large as a megabyte. Note that there are times that Netalyzr may have been under-detecting and/or underreporting the buffering, particularly on faster links; the Netalyzr group have been improving its buffer test.
Bittorrent Broadnets - with buffers of 20-40 packets (32-64 KB buffers). As soon as the common uplink saturated, everything went to heck
RAQM (REMOTE Active Queue
h2. More unsorted papers
Marina Thottan has been very kind to send me pointers to recent work on router buffer sizing. As you will see, there are good reasons to believe much conventional wisdom is far from the mark.
She is co-author of a survey paper that covers work done by different
groups on router buffer sizing. Here is a link to it:
Nick Mckeown @ Stanford has several talks on buffers for routers.
Here is the link to his web page:
The specific talks are on this page:
“Internet Routers: Past Present and Future.”
“Buffers: How we fell in love with them, and why we need a divorce.”
“Sizing Router Buffers”
“Network Processors and their memory”
“Designing Packet Buffers for Internet Routers”
“Memories for Internet Routers”