‹ Back
IETF 30 October 2013

Rough Guide to IETF 88: Scalability and Performance

Mat Ford
By Mat FordTechnology Programme Manager

The public policy world is full of discussions of appropriate (and inappropriate) management of bandwidth in the face of growing network usage. The Internet Engineering Task Force (IETF) and Internet Research Task Force (IRTF) have a number of efforts underway at IETF 88 in Vancouver next weekto explore and address more sophisticated ways to make use of available bandwidth, improve Internet performance, and otherwise get content to where it needs to be, efficiently.

Multipath TCP has been standardized in the IETF Multipath TCP working group and some recent exciting news was that Apple has implemented and deployed the protocol in iOS7. Ongoing refinements to the protocol to address security issues and advance the protocol on the Standards track will be discussed during the working group meeting in Vancouver.

Bufferbloat has been a hot topic among Internet transport geeks for some time now. It refers to the widespread presence of unmanaged and overly large buffers that exist in many network paths today, being common in consumer premises equipment (often referred to as CPE, including cable and DSL modems). These buffers go unnoticed most of the time but contribute significantly to Internet transaction latency when the bottleneck link is under load. The recently formed Active Queue Management (AQM) working group is chartered to develop algorithms for proactively managing queues (or buffers) in networking equipment in order to:

  1. help flow sources control their sending rates before the onset of necessary losses;
  2. help minimize delays for interactive applications; and
  3. 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.

One such application is browser-based real-time voice and video communication, otherwise known as RTCWeb. The RMCAT group is working on the especially thorny problem of how to handle congestion control for this highly sensitive application. Congestion control algorithms for interactive real time media may need to be quite different from the congestion control of TCP: for example, some applications can be more tolerant to loss than delay and jitter.

TCP is currently the Internet’s predominant transport protocol and the TCPM working grouphandles small TCP changes and minor extensions to TCP algorithms and protocol mechanisms. The group is working on a number of modifications to TCP that could significantly improve Internet performance as perceived by the average end user.

New or more general Internet transport related topics will be on the agenda for the catch-all Transport Area open meeting on Thursday from 9-1130AM, which will include a brief report of the recent Internet Society supported workshop on Reducing Internet Latency.

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 and the upcoming meeting will include interesting discussion of novel proposals for improving the algorithms used on the Internet to control congestion and thereby manage and optimize application performance for Internet users.

The Large-Scale Measurement of Broadband Performance (LMAP) group is working to standardize the LMAP measurement system for performance measurements of broadband access devices such as home and enterprise edge routers, personal computers, mobile devices, and set top boxes, whether wired or wireless.

LMAP, and other IETF efforts make use of metrics standardized by the IP Performance Metrics (IPPM) working group that has recently been re-chartered and had new chairs appointed. Newly adopted work items include bulk transfer capacity metrics and the definition of a registry of metric definitions.

Together, these efforts will help appropriately 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 88:

IEFT 88 Rough guide:

‹ Back

Disclaimer: Viewpoints expressed in this post are those of the author and may or may not reflect official Internet Society positions.

Related articles

Rough Guide to IETF 99: Scalability & Performance
Rough Guide to IETF 99: Scalability & Performance
Building Trust7 July 2017

Rough Guide to IETF 99: Scalability & Performance

In this post I'll highlight some of the Internet Engineering Task Force (IETF) and Internet Research Task Force (IRTF) groups...

Rough Guide to IETF 97: Scalability and Performance
Rough Guide to IETF 97: Scalability and Performance
Building Trust7 November 2016

Rough Guide to IETF 97: Scalability and Performance

In this first post of the Internet Society Rough Guide to IETF 97, I'll highlight some of the Internet Engineering...

Rough Guide to IETF 98: Scalability and Performance
Rough Guide to IETF 98: Scalability and Performance
Building Trust21 March 2017

Rough Guide to IETF 98: Scalability and Performance

In this Rough Guide to IETF 98 post I'll highlight some of the Internet Engineering Task Force (IETF) and Internet...

Join the conversation with Internet Society members around the world