Rate-Based Active Queue Management with Priority Classes for Better Video Transmission Jae Chung and Mark Claypool Computer Science Department Worcester.

Slides:



Advertisements
Similar presentations
IWSSC 2008 – Oct 3 rd 2008 Raffaello Secchi Scheduling TCP-Friendly Flows over a Satellite Network R. Secchi, A. Sathiaseelan, and G. Fairhurst Electronics.
Advertisements

Streaming Video over the Internet
WHITE – Achieving Fair Bandwidth Allocation with Priority Dropping Based on Round Trip Time Name : Choong-Soo Lee Advisors : Mark Claypool, Robert Kinicki.
CSIT560 Internet Infrastructure: Switches and Routers Active Queue Management Presented By: Gary Po, Henry Hui and Kenny Chong.
Top-Down Network Design Chapter Thirteen Optimizing Your Network Design Copyright 2010 Cisco Press & Priscilla Oppenheimer.
Bayesian Piggyback Control for Improving Real-Time Communication Quality Wei-Cheng Xiao 1 and Kuan-Ta Chen Institute of Information Science, Academia Sinica.
TCP/IP Protocol Suite 1 Copyright © The McGraw-Hill Companies, Inc. Permission required for reproduction or display. Chapter 25 Multimedia.
Receiver-driven Layered Multicast S. McCanne, V. Jacobsen and M. Vetterli University of Calif, Berkeley and Lawrence Berkeley National Laboratory SIGCOMM.
CS 268: Lecture 8 Router Support for Congestion Control Ion Stoica Computer Science Division Department of Electrical Engineering and Computer Sciences.
Chapter 10 Congestion Control in Data Networks1 Congestion Control in Data Networks and Internets COMP5416 Chapter 10.
Dynamic-CBT Better Performing Active Queue Management for Multimedia Networking Jae Chung and Mark Claypool Computer Science Department Worcester Polytechnic.
1 Measurements of the Congestion Responsiveness of Windows Streaming Media James Nichols, Mark Claypool, Robert Kinicki and Mingzhe Li Computer Science.
Presented by Santhi Priya Eda Vinutha Rumale.  Introduction  Approaches  Video Streaming Traffic Model  QOS in WiMAX  Video Traffic Classification.
Receiver-driven Layered Multicast S. McCanne, V. Jacobsen and M. Vetterli SIGCOMM 1996.
On Modeling Feedback Congestion Control Mechanism of TCP using Fluid Flow Approximation and Queuing Theory  Hisamatu Hiroyuki Department of Infomatics.
1 Estimating Shared Congestion Among Internet Paths Weidong Cui, Sridhar Machiraju Randy H. Katz, Ion Stoica Electrical Engineering and Computer Science.
Quality of Service Issues in Multi-Service Wireless Internet Links George Xylomenos and George C. Polyzos Department of Informatics Athens University of.
Application, Network and Link Layer Measurements of Streaming Video over a Wireless Campus Network Passive & Active Measurement Workshop 05 Boston, MA,
A Credit-based Home Access Point (CHAP) to Improve Application Performance on IEEE Networks Choong-Soo Lee, Mark Claypool and Robert Kinicki Worcester.
The War Between Mice and Elephants Presented By Eric Wang Liang Guo and Ibrahim Matta Boston University ICNP
Differentiated Services. Service Differentiation in the Internet Different applications have varying bandwidth, delay, and reliability requirements How.
1 Minseok Kwon and Sonia Fahmy Department of Computer Sciences Purdue University {kwonm, All our slides and papers.
ABE: Providing a Low Delay within Best Effort P. Hurley, M. Kara, J. Le Boudec, and P. Thiran ICA, Swiss Federal Institute of Technology, Lausanne, Switzerland.
AQM for Congestion Control1 A Study of Active Queue Management for Congestion Control Victor Firoiu Marty Borden.
MediaPlayer ™ vs. RealPlayer ™ A Comparison of Network Turbulence Mingzhe Li, Mark Claypool, Robert Kinicki CS Department Worcester Polytechnic Institute.
RAP: An End-to-End Rate-Based Congestion Control Mechanism for Realtime Streams in the Internet Reza Rejai, Mark Handley, Deborah Estrin U of Southern.
1 PV'2003, Nantes France, April 2003 Measurement of the Congestion Responsiveness of RealPlayer Streaming Video Over UDP Jae Chung, Mark Claypool, Yali.
1 Traffic Sensitive Quality of Service Controller Masters Thesis Submitted by :Abhishek Kumar Advisors: Prof Mark Claypool Prof Robert Kinicki Reader:
Adaptive Content-Aware Scaling for Improved Video Streaming. Avanish Tripathi Mark Claypool Presented by: Huahui Wu Worcester Polytechnic Institute
Traffic Sensitive Active Queue Management - Mark Claypool, Robert Kinicki, Abhishek Kumar Dept. of Computer Science Worcester Polytechnic Institute Presenter.
A Model for MPEG with Forward Error Correction (FEC) and TCP-Friendly Bandwidth Huahui Wu, Mark Claypool & Robert Kinicki Computer Science Department Worcester.
Dynamic-CBT and ChIPS – Router Support for Improved Multimedia Performance on the Internet Jae Chung and Mark Claypool Computer Science Department Worcester.
Stochastic Fair Traffic Management for Efficient and Robust IP Networking Jae Chung Airvana Inc. Chelmsford, MA Mark Claypool, Robert Kinicki WPI.
Dynamic-CBT and ChIPS - Router Support for Improved Multimedia Performance on the Internet Jae Chung and Mark Claypool Department of Computer Science WPI.
Better-Behaved Better- Performing Multimedia Networking Jae Chung and Mark Claypool (Avanish Tripathi) Computer Science Department Worcester Polytechnic.
ACN: Congestion Control1 Congestion Control and Resource Allocation.
The War Between Mice and Elephants By Liang Guo (Graduate Student) Ibrahim Matta (Professor) Boston University ICNP’2001 Presented By Preeti Phadnis.
Using Redundancy and Interleaving to Ameliorate the Effects of Packet Loss in a Video Stream Yali Zhu, Mark Claypool and Yanlin Liu Department of Computer.
Lightweight Active Router-Queue Management for Multimedia Networking M. Parris, K. Jeffay, and F.D. Smith Department of Computer Science University of.
A Study on Quality of Service Issues in Internet Telephony  IP Telephony – Applications and Services  Advantages and benefits of Voice over IP  Technical.
Better Behaved, Better Performing Multimedia Networking Jae Chung and Mark Claypool Computer Science Department Worcester Polytechnic Institute Proceedings.
Analysis of Active Queue Management Jae Chung and Mark Claypool Computer Science Department Worcester Polytechnic Institute Worcester, Massachusetts, USA.
February 7, 2003BU Computer Science Colloquium Crimson - Traffic Aware Active Queue Management Mark Claypool CS Department Worcester Polytechnic Institute.
Performance Evaluation of PISA and PI using NS simulations Presented by Brad Burres Yatin Manjrekar.
Using Interleaving to Ameliorate the Effects of Packet Loss in a Video Stream Mark Claypool and Yali Zhu Computer Science Department Worcester Polytechnic.
Quality of Service (QoS)
Top-Down Network Design Chapter Thirteen Optimizing Your Network Design Oppenheimer.
Link Scheduling & Queuing COS 461: Computer Networks
CIS679: DiffServ Model r Review of Last Lecture r 2-bit DiffServ architecture.
1 On Class-based Isolation of UDP, Short-lived and Long-lived TCP Flows by Selma Yilmaz Ibrahim Matta Computer Science Department Boston University.
Worcester Polytechnic Insitute, Worcester, MA, USA1 Traffic Sensitive Active Queue Management for Improved Multimedia Streaming Authors: Vishal Phirke,
The Impact of Active Queue Management on Multimedia Congestion Control Wu-chi Feng Ohio State University.
15744 Course Project1 Evaluation of Queue Management Algorithms Ningning Hu, Liu Ren, Jichuan Chang 30 April 2001.
Analysis of QoS Arjuna Mithra Sreenivasan. Objectives Explain the different queuing techniques. Describe factors affecting network voice quality. Analyse.
CS640: Introduction to Computer Networks Aditya Akella Lecture 20 - Queuing and Basics of QoS.
© 2006 Cisco Systems, Inc. All rights reserved. Optimizing Converged Cisco Networks (ONT) Module 3: Introduction to IP QoS.
Ch 6. Multimedia Networking Myungchul Kim
© 2006 Cisco Systems, Inc. All rights reserved. Optimizing Converged Cisco Networks (ONT) Module 4: Implement the DiffServ QoS Model.
CATNIP – Context Aware Transport/Network Internet Protocol Carey Williamson Qian Wu Department of Computer Science University of Calgary.
Adaptive Content-Aware Scaling for Improved Video Streaming. Avanish Tripathi Advisor: Mark Claypool Reader: Bob Kinicki.
© 2006 Cisco Systems, Inc. All rights reserved. Module 4: Implement the DiffServ QoS Model Lesson 4.6: Congestion Avoidance.
Instructor Materials Chapter 6: Quality of Service
Chapter 9 Optimizing Network Performance
Chapter 25 Multimedia TCP/IP Protocol Suite
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 6: Quality of Service Connecting Networks.
Traffic Sensitive Active Queue Management
A Selective Retransmission Protocol for Multimedia on the Internet
Providing QoS through Active Domain Management
Congestion Control for Streaming Media
Adjusting Forward Error Correction for TCP- Friendly Streaming MPEG
Presentation transcript:

Rate-Based Active Queue Management with Priority Classes for Better Video Transmission Jae Chung and Mark Claypool Computer Science Department Worcester Polytechnic Institute Worcester, Massachusetts, USA

ISCC02 WPI Rb-RIO #2 Streaming Media Applications Can use TCP –Hides network information –Difficult to use media scaling upon –Hard to control transmission timing –Higher jitter due to traffic burst Often use UDP –Control the flow as desired Responsiveness, Fairness –Can tolerate some frame losses Trickier for Video Streaming

ISCC02 WPI Rb-RIO #3 Difficulty of Video Streaming Inter-frame dependencies (Compression) Large frames: fragmented in IP layer IBPBBBI A Sample MPEG Stream I BPBBB

ISCC02 WPI Rb-RIO #4 Outline Introduction Approaches Rb-RIO Experiments Results Summary

ISCC02 WPI Rb-RIO #5 Approaches to The Problem Application Support –Redundancy and Error Correction –Media Scaling Network Architectural Support –Integrated Services (IntServ) –Differentiated Services (DiffServ) Router Support –Resource Reservation (WFQ, …) –Priority Class-Based Queue Management

ISCC02 WPI Rb-RIO #6 Basic Idea: Avoid Domino Frame Loss Using Priority Queuing Queue supports 3 priority class –Drop all the lowest class packets before starting to drop the next class packets MPEG Video Frame Mapping –Map I-, P- and B-frame packets to High, Mid and Low priority classes TCP Mapping –Randomly map TCP packets to the 3 classes (standalone better-than-best-effort Queue)

ISCC02 WPI Rb-RIO #7 AQM Support for Priority-Classes … … RED RIO Average Queue Size Based … … Rb-RED Rb-RIO Rate Based - Detect Congestion - Determine Degree - Notify Sources Active Queue Management

ISCC02 WPI Rb-RIO #8 Rb-RED Components –Limit:Queue Limit –Q:Queue Size –EAR:Estimated Arrival Rate –SR:Service Rate Drop Logic (for incoming packets) if (Q > Limit) Drop else if (EAR > SR) Drop with P = (EAR – SR) / EAR SR EAR

ISCC02 WPI Rb-RIO #9 Rb-RIO Drop Logic if (Q > Limit) Drop else if (EAR high > SR) if (L-class or M-class) Drop if (H-class) Drop with P = (EAR high – SR) / EAR high else if (EAR high + EAR mid > SR) if (L-class) Drop if (M-class) Drop with P = (EAR high + EAR mid – SR) / EAR mid else if (EAR > SR) if (L-class) Drop with P = (EAR – SR) / EAR low SR EAR EAR low EAR mid EAR high SR EAR EAR low EAR mid EAR high EAR EAR low EAR mid EAR high SR

ISCC02 WPI Rb-RIO #10 Rb-RIO Mappings MPEG Video –I-frame packets= H-class –P-frame packets= M-class –B-frame packets = L-class TCP –Randomly map incoming TCP packets –H-class : M-class : L-class = x : y : z

ISCC02 WPI Rb-RIO #11 Outline Introduction Approaches Rb-RIO Experiments Results Summary

ISCC02 WPI Rb-RIO #12 Experiments Implement Rb-RIO in NS General Settings –Network Packet Size = 1 KB –MPEG Frame Size (from a news clip) I, P, B = 11, 8, 2 (KB) –Rb-RIO Priority Class Mapping for TCP H : M : L = 7 : 3 : 0 Compare Rb-RIO with RED, Drop-Tail –Using TCP, MPEG and MIX (TCP + MPEG)

ISCC02 WPI Rb-RIO #13 n1-n2:q_size= 100 qweight= RED:max_th= 60 min_th= 20 max_p= 0.1 Rb-RIO: interval= 1ms ear_k= 0.5 aggr_p= Flows 20 Flows Simulation Time (Seconds) s59 s60 s1 s2 n1 r59 r60 r1 r2 n2 25Mbps, 20ms 25Mbps, 5ms packet_size = 1kb FTP-TCP EXP TCP: Setup

ISCC02 WPI Rb-RIO #14 EXP TCP: Summary Comparing Rb-RIO with Drop-Tail and RED (well-configured) –Throughput:All compatible –Drop Rate:All compatible –Fairness:RED and Rb-RIO are compatible and fairer than Drop-Tail

ISCC02 WPI Rb-RIO #15 EXP MPEG: Setup n1-n2:q_size= 100 qweight= RED:max_th= 60 min_th= 20 max_p= 0.1 Rb-RIO: interval= 1ms ear_k= 0.5 aggr_p= Flows 20 Flows Simulation Time (Seconds) s59 s60 s1 s2 n1 r59 r60 r1 r2 n2 25Mbps, 20ms 25Mbps, 5ms packet_size = 1kb Responsive MPEG Unresponsive MPEG

ISCC02 WPI Rb-RIO #16 EXP MPEG: Pkt & Frm Statistics ThroughputPkt Loss %Frm Loss % Deco Loss % Drop-Tail19.82 Mbps RED18.65 Mbps Rb-RIO20.81 Mbps Responsive MPEG Unresponsive MPEG ThroughputPkt Loss %Frm Loss % Deco Loss % Drop-Tail14.27 Mbps RED15.18 Mbps Rb-RIO 9.70 Mbps

ISCC02 WPI Rb-RIO #17 EXP MPEG: Decodable Frames (Responsive MPEG)

ISCC02 WPI Rb-RIO #18 EXP MPEG: Decodable Frames (Unresponsive MPEG)

ISCC02 WPI Rb-RIO #19 EXP MPEG: Fairness

ISCC02 WPI Rb-RIO #20 EXP MPEG: Summary Rb-RIO improves video transmission over Drop-Tail and RED –Effective use of network bandwidth –Smoother video streams (less jitter) Improves fairness among video streams –Same kind of streams –Different kind of streams

ISCC02 WPI Rb-RIO #21 EXP MIX: Setup n1-n2:q_size= 100 qweight= RED:max_th= 60 min_th= 20 max_p= 0.1 Rb-RIO: interval= 1ms ear_k= 0.5 aggr_p= Flows Simulation Time (Seconds) s59 s60 s1 s2 n1 r59 r60 r1 r2 n2 25Mbps, 20ms 25Mbps, 5ms packet_size = 1kb Responsive MPEG Unresponsive MPEG 20 Flows FTP-TCP

ISCC02 WPI Rb-RIO #22 EXP MIX: Fairness

ISCC02 WPI Rb-RIO #23 EXP MIX: Summary Rb-RIO improves video transmission over Drop-Tail and RED as in EXP MPEG Protects TCP from other flows that are unresponsive or response differently to network congestion

ISCC02 WPI Rb-RIO #24 Summary Rb-RIO behaves compatibly with well- configured RED on TCP flows Offers efficient and smoother video transmission Improves fairness among video streams Protects TCP using configurable TCP mapping ratio

ISCC02 WPI Rb-RIO #25 Future Work Thorough evaluation of Rate-Based AQM mechanism (with TCP) Try and evaluate Rb-RIO with other video streams (H.263+, RealVideo) Design an AQM mechanism that can control delay for interactive multimedia streaming