Flow Rate Fairness Many slides are borrowed from Bob Briscoe

Slides:



Advertisements
Similar presentations
Congestion Control and Fairness Models Nick Feamster CS 4251 Computer Networking II Spring 2008.
Advertisements

20.1 Chapter 20 Network Layer: Internet Protocol Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display.
IPv4 - The Internet Protocol Version 4
1 IP - The Internet Protocol Relates to Lab 2. A module on the Internet Protocol.
Transport Layer – TCP (Part1) Dr. Sanjay P. Ahuja, Ph.D. Fidelity National Financial Distinguished Professor of CIS School of Computing, UNF.
Congestion Control An Overview -Jyothi Guntaka. Congestion  What is congestion ?  The aggregate demand for network resources exceeds the available capacity.
Explicit Congestion Notification (ECN) Qi (Gill) Wang CISC 856 – TCP/IP, Fall 2012 Special thanks to: Dr. Paul Amer Guna Ranjan, Justin.
1 Congestion Control. Transport Layer3-2 Principles of Congestion Control Congestion: r informally: “too many sources sending too much data too fast for.
1 Internet Networking Spring 2003 Tutorial 11 Explicit Congestion Notification (RFC 3168) Limited Transmit (RFC 3042)
1 Internet Networking Spring 2003 Tutorial 11 Explicit Congestion Notification (RFC 3168)
1 Spring Semester 2007, Dept. of Computer Science, Technion Internet Networking recitation #8 Explicit Congestion Notification (RFC 3168) Limited Transmit.
HUAWEI TECHNOLOGIES CO., LTD. Huawei Confidential Security Level: Slide title :40-47pt Slide subtitle :26-30pt Color::white Corporate Font : FrutigerNext.
Transport Layer 4 2: Transport Layer 4.
These materials are licensed under the Creative Commons Attribution-Noncommercial 3.0 Unported license (
Introduction to Networks CS587x Lecture 1 Department of Computer Science Iowa State University.
1 Congestion Control Computer Networks. 2 Where are we?
CS 4396 Computer Networks Lab
Network Management has always been and always will be essential to the Internet Testimony of George Ou Former Network Engineer FCC.
Explicit Congestion Notification (ECN) RFC 3168
1 Transport Layer: Basics Outline Intro to transport UDP Congestion control basics.
Providing QoS in IP Networks
Protocols and layering Network protocols and software Layered protocol suites The OSI 7 layer model Common network design issues and solutions.
Example DLL Protocols 1. High-Level Data Link Control (HDLC).
Lecture (2).
Internet Networking recitation #9
Topics discussed in this section:
Chapter 3 outline 3.1 transport-layer services
IP - The Internet Protocol
Top-Down Network Design Chapter Thirteen Optimizing Your Network Design Copyright 2010 Cisco Press & Priscilla Oppenheimer.
5. End-to-end protocols (part 1)
Chapter 6 Congestion Avoidance
Seminar report on IPv4 & IPv6
Chapter 3 outline 3.1 Transport-layer services
Congestion Control and Resource Allocation
IP - The Internet Protocol
Net431:advanced net services
TASK 4 Guideline.
Congestion Control, Internet transport protocols: udp
Packet Sniffing.
Net 431: ADVANCED COMPUTER NETWORKS
Transport Layer Unit 5.
TCP, XCP and Fair Queueing
Chapter 20 Network Layer: Internet Protocol
IP - The Internet Protocol
Dr. John P. Abraham Professor UTPA
Chapter 20 Network Layer: Internet Protocol
Dr. John P. Abraham Professor UTRGV, EDINBURG, TX
IT351: Mobile & Wireless Computing
Congestion Control (from Chapter 05)
COS 461: Computer Networks
IP - The Internet Protocol
COMP/ELEC 429/556 Introduction to Computer Networks
Dr. John P. Abraham Professor UTPA
Internet Networking recitation #10
Net 323 D: Networks Protocols
IP - The Internet Protocol
Chapter 11. Frame Relay Background Frame Relay Protocol Architecture
Building A Network: Cost Effective Resource Sharing
Congestion Control (from Chapter 05)
Congestion Control (from Chapter 05)
was not invented by Al Gore…
Transport Layer: Congestion Control
Congestion Control (from Chapter 05)
IPv4 Addressing By, Ishivinder Singh( ) Sharan Patil ( )
Congestion Control (from Chapter 05)
Congestion Control and Resource Allocation
IP - The Internet Protocol
NET 323D: Networks Protocols
Transport Layer 9/22/2019.
Lecture 6, Computer Networks (198:552)
Presentation transcript:

Flow Rate Fairness Many slides are borrowed from Bob Briscoe http://www.bobbriscoe.net/projects/refb/ Presented by: Yang Guan March 25, 2010 CISC 856: TCP/IP & Upper Layer Protocols

Resource Sharing in the Internet The Internet is based on a simple premise: Sharing communication links are more efficient than dedicated channels The primary sharing algorithm is built into Transmission Control Protocol (TCP) TCP provides mechanisms to guard people how to share Internet capacity politely For economical reasons: dedicated is expensive For channel utilization: a physical channel may remain idle if only one user We share local area networks at work and neighborhood links from home. Indeed, a multi-gigabit backbone cable is shared among thousands of folks surfing the Web, downloading videos, and talking on Internet phones. That is congestion control: TCP constantly does congestion detection that lows its 2/17/2019

How TCP shares the Internet The protocol allows you to seem to be polite TCP constantly increases transmission rate if it can Until it sees some sign of congestion, TCP politely reduces bit rate TCP-friendly if new protocol does not consume more bit rate than TCP does. 2/17/2019

TCP-Friendliness TCP is even used as a standard For applications that do not utilize TCP in transport layer, they are called TCP-friendly if they consume about the same data rate as TCP does. 2/17/2019

Does TCP make the world perfect? The answer is of course NO! Methods to circumvent TCP-friendliness rules: Running multiple TCP sessions Running each TCP session for long time It is really the application software that determines how to share the Internet fairly So,does TCP make the world perfect? When I ask this question, you know the answer is NO. Otherwise this is just the end of my talk. It is really the end application software that determine how to share the Internet fairly It is not right to solely blame P2P software, the first Web browser opens 4 TCP connections. The TCP itself is problematic. 2/17/2019

Fig 1. TCP overlooks users’ activity over time [4] What does TCP overlook? The plot visualizes the shares of the ISP's 10Mbps link taken by 10 sample users. Those active at any one moment take equal rates, regardless of how active each user is over time. In general, TCP gives 20 shares of bottleneck capacity to the heavy users and only 4 to the light users. Fig 1. TCP overlooks users’ activity over time [4] 2/17/2019

What does TCP overlook? (cont’d) When there are only 20 active downloading, bandwidth available to each light user can be as low as 20 kilobits, which is worse than dial-up We see Simply upgrade physical link does not solve the fairness problem. Fig 2. TCP overlooks multiple TCP instances [4] 2/17/2019

Rethink: What is fair? Equal flow rate? It is not about how much a TCP consumes It is about how much a TCP can affect others 2/17/2019

Rethink: What is fair? (cont’d) How to measure the effect on others? Congestion volume: the amount of data that is sent during network congestion 2/17/2019

Rethink: What is fair? (cont’d) Network neutrality debate, whether ISPs should limit the usage of certain applications. The same as discussing with fox on how to set up barrier on a chicken coop…… Fig 3. Different TCP sharing schemes [4] 2/17/2019

Rethink: What is fair? (cont’d) Fair is faster: Light browsing goes blisteringly faster Heavy downloading is not obviously prolonged 2/17/2019

Problems with TCP Congestion is only detected and managed solely by computers at the edge ISPs cannot set congestion limits The few ruin the life of the many Massive capacity is required But poor incentive to invest in capacity To make router as simple as possible. 2/17/2019

A New TCP Routine Parameterize TCP with weight The key is Behave like 12 TCP flows, or Behave like 0.25 of a TCP flow The key is High weights for light interactive usage (web surfing) Low weights for heavy usage (movie downloads) 2/17/2019

A New TCP Routine (cont’d) Whenever congestion happens Higher weighted TCP goes much faster Lower weighted TCP expands back to fast rate afterwards 2/17/2019

A New TCP Routine (cont’d) On today’s Internet, the balance of weights is the wrong way around How to persuade people to reasonably choose weights? We should limit people by the effects they have on others—the incremental cost of their usage Congestion volume: the volume of data sent during congestion 2/17/2019

A New TCP Routine (cont’d) Solution: ISPs provide a monthly congestion-volume allowance (CVA) High weights TCPs consumes CVA while low weights ones doesn’t Heavy usage does not consume CVA since weights are set to be low Light intensive usage does not consume too much CVA due to short lifetime 2/17/2019

congestion policer – one example: per-user policer NB NA R1 S1 overdraft non-interactive long flows (e.g. P2P, ftp) congestion volume allowance interactive short flows (e.g. Web, IM)

Making Congestion Visible to Network Layer Why? Healthy supply of bandwidth Reroute data around congested links Costumers draw down the limited allowances if congestion can not be avoided. Currently, only the router that drops a packet knows the drop 9 8 7 6 5 3 2/17/2019

First Step: ECN Explicit Congestion Notification Standardized into TCP/IP in 2001 ECN allows end-to-end notification of network congestion without dropping packets[3] Routers set CE (Congestion Experience) bit when the average queue length exceeds configured threshold levels. Receivers feedback congestion information back to senders Traditionally, TCP/IP networks signal congestion by dropping packets. When ECN is successfully negotiated, an ECN-aware router may set a bit in the IP header instead of dropping a packet in order to signal the beginning of congestion. The receiver of the packet echoes the congestion indication to the sender, which must react as though a packet drop were detected. implemented by most router vendors – very lightweight mechanism but rarely turned on by operators (yet) – mexican stand-off with OS vendors 2/17/2019

First Step: ECN (cont’d) VER HLEN Service Type Total length Identification Flag Fragmentation offset TTL Protocol Header checksum Source IP address Destination IP address Fig 4. IPv4 header format 2/17/2019

First Step: ECN (cont’d) feedback 2 3 4 5 6 7 8 9 NB NA R S 8 7 6 5 4 3 2 Fig 5. ECN mechanism [5] 2/17/2019

Second Step: re-feedback Fig 6. Re-feedback mechanism [4] 2/17/2019

Conclusion Ready to be implemented as ECN has been included into TCP/IP Sticks to the Internet e2e principle Makes congestion visible to the networks in the middle 2/17/2019

References [1] Bob Briscoe (BT), Illustrations by QuickHoney, A Fairer, Faster Internet Protocol, IEEE Spectrum, Dec 2008 pp38-43 [2] B. Briscoe. Flow rate fairness: Dismantling a religion. Computer Communications Review, 37(2):63–74, Apr. 2007. [3] Explicit Congestion Notification, http://en.wikipedia.org/wiki/Explicit_Congestion_Notification [4] Bob Briscoe, Internet: Fairer is Faster, BT White Paper [5] Bob Briscoe, et al, Policing congestion response in an internetwork using re-feedback, Sigcomm 2005 2/17/2019

Questions Thanks 2/17/2019