[AusNOG] Bufferbloat - Queue management vs Buffer filling

[Andrew Cox]

>From Jim Gettys of W3C/GNOME/IETF/OLPC fame:
https://gettys.wordpress.com/category/bufferbloat/

Figured some of you might find these articles a heavy, but interesting read.
*

Bufferbloat is existence of excessively large (bloated) buffers into
systems, particularly network communication systems.

Systems suffering from bufferbloat will have bad latency under load under
some or all circumstances, depending on if and where the bottleneck in the
communication’s path exists. Bufferbloat encourages congestion of networks;
bufferbloat destroys congestion avoidance in transport protocols such as
HTTP, TCP, Bittorrent, etc. Network congestion avoidance algorithms depend
upon timely packet drops or ECN; bloated buffers violate this design
presumption. Without active queue management, these bloated buffers will
fill, and stay full.
*
What I'm interested to know is if any of you have encountered/investigated
this yourself or would care to comment further on the validity/relevance of
such in an ISP environment approaching NBN speeds.

More info about some of the Netalyzer reports here:
http://www.icir.org/christian/publications/2010-imc-netalyzr.pdf

Kind Regards,
Andrew Cox
AccessPlus - Head Network Administrator
Ph: 1300 739 822 (7am - 12 midnight AEST)
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ausnog.net/pipermail/ausnog/attachments/20110116/63d3d110/attachment.html>