Open Internet Standards Technology

Rough Guide to IETF 89: Internet Scalability and Performance

In this post I’ll shine a light on some of the Internet Engineering Task Force (IETF) and Internet Research Task Force (IRTF) efforts underway to explore and address more sophisticated ways to use available bandwidth, improve Internet performance, and otherwise efficiently get content to where it needs to be. These groups will all be meeting as part of the IETF 89 meeting in London, England, next week.

There are two birds-of-a-feather (BoF) meetings worth highlighting. The Tunneling Compressed Multiplexed Traffic Flows (TCMTF) BoF is looking to update existing work on optimising real-time packet flows for bandwidth savings. One of the proponents of this work, Jose Saldana, wrote about the group’s motivations and goals in the IETF Journal recently:

The other BoF meeting of interest is Transport Services (TAPS). This meeting will feature a more architectural discussion about the services a transport API is required to provide based on the kinds of things application developers are now using UDP and TCP for. The group will then analyse how these required transport services could be implemented using existing protocols, and how to validate that a specific transport service can be supported on a given path.

TCP is currently the Internet’s predominant transport protocol and the TCPM Working Group handles small TCP changes and minor extensions to TCP algorithms and protocol mechanisms. In addition to continuing its work on TCP modifications that could significantly improve Internet performance as perceived by the average end user, the TCPM meeting in London will include a ‘mini-BoF’ on TCPcrypt. This is responsive to the threat of pervasive surveillance and, while TCPcrypt has been around for some time and faces substantial deployment challenges, the newly exposed threat model has given rise to renewed interest in exploring the potential for TCPcrypt deployment.

The Active Queue Management (AQM) Working Group is continuing its work to develop algorithms for proactively managing queues (or buffers) in networking equipment in order to:

  • help flow sources control their sending rates before the onset of necessary losses;
  • help minimize delays for interactive applications; and
  • help protect flows from negative impacts of other more aggressive or misbehaving flows.

Once deployed, AQM and packet scheduling algorithms can have a huge impact on the responsiveness, resilience, and usability of the Internet for diverse applications.

The Internet is a shared medium for communication, which requires all end systems to abide by certain rules to prevent ‘congestion collapse’ of the network. The Internet Congestion Control Research Group (ICCRG) is the IRTF home for work on congestion control. The upcoming meeting (and the AQM working group meeting mentioned above) will include presentation and discussion of the variant of the PIE algorithm adopted by CableLabs for implementation in DOCSIS 3.1 cable modems.

The Large-Scale Measurement of Broadband Performance (LMAP) and IP Performance Metrics (IPPM) Working Groups will continue their work to define metrics and a measurement system to inform the public policy debate, and educate end users about the performance of ISP offerings in their marketplace by providing standardized metrics and a framework for measuring network performance.

Related Working Groups and BoFs at IETF 89:

To follow along as we dole out this series of Rough Guide to IETF blog posts, follow us on the Internet Technology Matters blog, Twitter, Facebook, Google+, via RSS, or see