^ Top

NANOG Meeting Presentation Abstract

Research Forum: Sizing Router Buffers
Meeting: NANOG32
Date / Time: 2004-10-18 4:05pm - 5:00pm
Room: Grand Ballroom
Presenters: Speakers:

Guido Appenzeller, Stanford University

Guido Appenzeller is currently finishing his Ph.D. at Stanford University, where his thesis topic is the sizing of router buffers with Prof. Nick McKeown. Guido is also founder and CTO of Voltage Security, a Palo Alto-based startup pioneering identity-based encryption technology. He was recently named to the MIT Technology Review\'s prestigious TR100 list of top young innovators.
Abstract: All Internet routers contain buffers to hold packets during times of congestion. Today, the size of the buffers is determined by the dynamics of TCP\'s congestion control algorithm. In particular, the goal is to make sure that when a link is congested, it is busy 100% of the time; which is equivalent to making sure its buffer never goes empty. A widely used rule-of-thumb states that each link needs a buffer of size B = RTT X C, where RTT is the average round-trip time of a flow passing across the link, and C is the data rate of the link. For example, a 10Gb/s router linecard needs approximately 250ms X 10Gb/s = 2.5Gbits of buffers; and the amount of buffering grows linearly with the line-rate. Such large buffers are challenging for router manufacturers, who must use large, slow, off-chip DRAMs. And queueing delays can be long, have high variance, and may destabilize the congestion control algorithms.



In this talk we argue that the rule-of-thumb B = RTT X C is now outdated and incorrect for backbone routers. This is because of the large number of flows (TCP connections) multiplexed together on a single backbone link. Using theory, simulation and experiments on a network of real routers, we show that a link with n flows requires no more than B =(RTT X C) / sqrt{n}, for long-lived or short-lived TCP flows. The consequences on router design are enormous: A 2.5Gb/s link carrying 10,000 flows could reduce its buffers by 99% with negligible difference in throughput; and a 10Gb/s link carrying 50,000 flows requires only 10Mbits of buffering, which can easily be implemented using fast, on-chip SRAM.



We also present recent data on how a live network with real traffic performs with reduced router buffers.
Files: pdfGuido Appenzeller Presentation(PDF)
pptGuido Appenzeller Presentation (Animation)(PPT)
youtubeResearch Forum: Sizing Router Buffers
Sponsors: None.

Back to NANOG32 agenda.

NANOG32 Abstracts

  • Botnets
    Speakers:
    John Kristoff, Northwestern University;
  • BGP—The Movie
    Speakers:
    Geoff HustonAPNIC; .
    George MichaelsonAPNIC; .
    Philip SmithCisco Systems; .
  • BGP—The Movie
    Speakers:
    Geoff HustonAPNIC; .
    George MichaelsonAPNIC; .
    Philip SmithCisco Systems; .
  • BGP—The Movie
    Speakers:
    Geoff HustonAPNIC; .
    George MichaelsonAPNIC; .
    Philip SmithCisco Systems; .
  • Life and Times of J-Root
    Speakers:
    Piet BarberVerisign; .
    Matt LarsonVerisign; .
    Mark KostersVerisign; .
    Pete ToscanoVerisign; .
  • Life and Times of J-Root
    Speakers:
    Piet BarberVerisign; .
    Matt LarsonVerisign; .
    Mark KostersVerisign; .
    Pete ToscanoVerisign; .
  • Life and Times of J-Root
    Speakers:
    Piet BarberVerisign; .
    Matt LarsonVerisign; .
    Mark KostersVerisign; .
    Pete ToscanoVerisign; .
  • Life and Times of J-Root
    Speakers:
    Piet BarberVerisign; .
    Matt LarsonVerisign; .
    Mark KostersVerisign; .
    Pete ToscanoVerisign; .

 
^ Back to Top