Scalable Reliable Multicast in Wide Area Networks Sneha Kumar Kasera Department of Computer Science University of Massachusetts, Amherst.

Slides:



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

Internet Measurement Conference 2003 Source-Level IP Packet Bursts: Causes and Effects Hao Jiang Constantinos Dovrolis (hjiang,
Push Technology Humie Leung Annabelle Huo. Introduction Push technology is a set of technologies used to send information to a client without the client.
Jump to first page A. Patwardhan, CSE Digital Fountains Main Ideas : n Distribution of bulk data n Reliable multicast, broadcast n Ideal digital.
Lava: A Reality Check of Network Coding in Peer-to-Peer Live Streaming Mea Wang, Baochun Li Department of Electrical and Computer Engineering University.
BZUPAGES.COM 1 User Datagram Protocol - UDP RFC 768, Protocol 17 Provides unreliable, connectionless on top of IP Minimal overhead, high performance –No.
Optimizing Buffer Management for Reliable Multicast Zhen Xiao AT&T Labs – Research Joint work with Ken Birman and Robbert van Renesse.
Congestion Control Created by M Bateman, A Ruddle & C Allison As part of the TCP View project.
Router Buffer Sizing and Reliability Challenges in Multicast Aditya Akella 02/28.
L-21 Multicast. L -15; © Srinivasan Seshan, Overview What/Why Multicast IP Multicast Service Basics Multicast Routing Basics DVMRP Overlay.
1 Improving the Performance of Distributed Applications Using Active Networks Mohamed M. Hefeeda 4/28/1999.
Satellite Multicast for Web Applications Hilmar Linder University of Salzburg/Austria.
1-1 CMPE 259 Sensor Networks Katia Obraczka Winter 2005 Transport Protocols.
A loss detection Service for Active Reliable Multicast Protocols Moufida MAIMOUR & C. D. PHAM INRIA-RESO RESAM UCB-Lyon – ENS Lyon INC’02, Plymouth Tuesday,
Network Multicast Prakash Linga. Last Class COReL: Algorithm for totally-ordered multicast in an asynchronous environment, in face of network partitions.
1 Chapter 3 Transport Layer. 2 Chapter 3 outline 3.1 Transport-layer services 3.2 Multiplexing and demultiplexing 3.3 Connectionless transport: UDP 3.4.
Data Communications Architecture Models. What is a Protocol? For two entities to communicate successfully, they must “speak the same language”. What is.
Resilient Multicast Support for Continuous-Media Applications X. Xu, A. Myers, H. Zhang and R. Yavatkar CMU and Intel Corp NOSSDAV, 1997.
Multicast Networking 2 References Multicast Networking and Applications Miller, C. Kenneth Addison-Wesley, 1999 Computer Networking:
1 A Comparison of Mechanisms for Improving TCP Performance over Wireless Links Course : CS898T Instructor : Dr.Chang - Swapna Sunkara.
3-1 Transport services and protocols r provide logical communication between app processes running on different hosts r transport protocols run in end.
An Active Reliable Multicast Framework for the Grids M. Maimour & C. Pham ICCS 2002, Amsterdam Network Support and Services for Computational Grids Sunday,
Error Checking continued. Network Layers in Action Each layer in the OSI Model will add header information that pertains to that specific protocol. On.
Multicast Transport Protocols: A Survey and Taxonomy Author: Katia Obraczka University of Southern California Presenter: Venkatesh Prabhakar.
Capacity of Ad Hoc Networks Quality of Wireless links Physical Layer Issues The Channel Capacity Path Loss Model and Signal Degradation MAC for.
CSE679: Multicast and Multimedia r Basics r Addressing r Routing r Hierarchical multicast r QoS multicast.
Receiver-driven Layered Multicast Paper by- Steven McCanne, Van Jacobson and Martin Vetterli – ACM SIGCOMM 1996 Presented By – Manoj Sivakumar.
Multicast Congestion Control in the Internet: Fairness and Scalability
Lect3..ppt - 09/12/04 CIS 4100 Systems Performance and Evaluation Lecture 3 by Zornitza Genova Prodanoff.
Spring 2000Nitin BahadurAdvanced Computer Networks A Comparison of Mechanisms for Improving TCP Performance over Wireless Links By: Hari B., Venkata P.
Advanced Network Architecture Research Group 2001/11/149 th International Conference on Network Protocols Scalable Socket Buffer Tuning for High-Performance.
3: Transport Layer3b-1 Principles of Congestion Control Congestion: r informally: “too many sources sending too much data too fast for network to handle”
Transport Layer3-1 Chapter 3 outline r 3.1 Transport-layer services r 3.2 Multiplexing and demultiplexing r 3.3 Connectionless transport: UDP r 3.4 Principles.
Dec 4, 2007 Reliable Multicast Group Neelofer T. CMSC 621.
ARQ Mechanisms Rudra Dutta ECE/CSC Fall 2010, Section 001, 601.
Computer Networks Performance Metrics. Performance Metrics Outline Generic Performance Metrics Network performance Measures Components of Hop and End-to-End.
Wireless TCP Prasun Dewan Department of Computer Science University of North Carolina
7/26/ Design and Implementation of a Simple Totally-Ordered Reliable Multicast Protocol in Java.
Advanced Network Architecture Research Group 2001/11/74 th Asia-Pacific Symposium on Information and Telecommunication Technologies Design and Implementation.
Transport Layer 3-1 Chapter 3 Transport Layer Computer Networking: A Top Down Approach 6 th edition Jim Kurose, Keith Ross Addison-Wesley March
1 Flow and Congestion Control for Reliable Multicast Communication In Wide-Area Networks Supratik Bhattacharyya Department of Computer Science University.
HighSpeed TCP for High Bandwidth-Delay Product Networks Raj Kettimuthu.
2007/03/26OPLAB, NTUIM1 A Proactive Tree Recovery Mechanism for Resilient Overlay Network Networking, IEEE/ACM Transactions on Volume 15, Issue 1, Feb.
1 Flow and Congestion Control for Reliable Multicast Communication In Wide-Area Networks A Doctoral Dissertation By Supratik Bhattacharyya.
TCP with Variance Control for Multihop IEEE Wireless Networks Jiwei Chen, Mario Gerla, Yeng-zhong Lee.
CIS679: Multicast and Multimedia (more) r Review of Last Lecture r More about Multicast.
Page 1 The department of Information & Communications Engineering Dong-uk, kim A Survey of Packet Loss Recovery Techniques for Streaming.
Video Multicast over the Internet Presented by: Liang-Yuh Wu Lung-Yuan Wu Hao-Hsiang Ku 12 / 6 / 2001 Bell Lab. And Georgia Institute of Technologies IEEE.
Flow and Congestion Control for Reliable Multicast Communication In Wide-Area Networks Supratik Bhattacharyya Department of Computer Science University.
Department of Electronic Engineering City University of Hong Kong EE3900 Computer Networks Protocols and Architecture Slide 1 Use of Standard Protocols.
Reliable Multicast Routing for Software-Defined Networks.
Advance Computer Networks Lecture#09 & 10 Instructor: Engr. Muhammad Mateen Yaqoob.
EE689 Lecture 13 Review of Last Lecture Reliable Multicast.
TCP OVER ADHOC NETWORK. TCP Basics TCP (Transmission Control Protocol) was designed to provide reliable end-to-end delivery of data over unreliable networks.
Security for Broadcast Network
Shivkumar Kalyanaraman Rensselaer Polytechnic Institute 1 ECSE-6600: Internet Protocols Informal Quiz #09: SOLUTIONS Shivkumar Kalyanaraman: GOOGLE: “Shiv.
CS640: Introduction to Computer Networks Aditya Akella Lecture 15 TCP Congestion Control.
1 Highlight 1: AER/NCA. 2 Active Multicast Repair Services Receiver Sender Conventional Routers Repair latency is a complete round trip time Link causing.
Reliable Adaptive Lightweight Multicast Protocol Ken Tang, Scalable Network Technologies Katia Obraczka, UC Santa Cruz Sung-Ju Lee, Hewlett-Packard Laboratories.
Transmission Control Protocol (TCP) TCP Flow Control and Congestion Control CS 60008: Internet Architecture and Protocols Department of CSE, IIT Kharagpur.
Computer Networking Lecture 16 – Reliable Transport.
Accelerating Peer-to-Peer Networks for Video Streaming
CMPE 252A: Computer Networks
Chapter 3 outline 3.1 transport-layer services
COMP 431 Internet Services & Protocols
CIS, University of Delaware
Reliable Multicast Group
Tarun Banka Department of Computer Science Colorado State University
Transport Layer: Congestion Control
Error Checking continued
Presentation transcript:

Scalable Reliable Multicast in Wide Area Networks Sneha Kumar Kasera Department of Computer Science University of Massachusetts, Amherst

Why Multicast ? multiple unicast broadcastmulticast one sender three receivers

Why Reliable Multicast ? applications one-to-many file transfer information updates (e.g., stock quote, web cache updates) shared whiteboard multicast lossy network

Goal design, evaluate multicast loss recovery approaches that make efficient use of end-host, network resources scale to several thousand receivers spanning wide area networks

Feedback Implosion NAK implosion ? NAK suppression (using timers) NAK aggregation (by building hierarchy) pktACK sender receiver receivers pkt ACK pkt NAK loss problem: ACK implosionsolution: use NAKs

sender loss original transmission loss pkt lost retransmission unicast multicast

Problem of Retransmission Scoping if same channel for retransmissions, retransmissions go everywhere how to shield receiver from loss recovery due to other receivers ? sender loss original transmission loss retransmission

Loss Recovery Burden when #receivers large, each pkt lost at some rcvr with high probability sender retransmits almost all pkts several times how to share burden of loss recovery ? sender loss pkt 1 pkt 4 pkt 2 pkt 3 loss retransmits pkts 1, 2, 3, 4

scoping retransmissions using multiple multicast channels server-based local recovery performance benefits resource requirements “active” repair services signaling for locating, invoking, revoking services router support Thesis Contributions

scoping retransmissions using multiple multicast channels server-based local recovery performance benefits resource requirements “active” repair services signaling for locating, invoking, revoking services router support summary and future directions Overview

one channel for original transmissions, A orig additional channels for retransmissions, pkt k sent on A k on detecting loss of pkt k, receiver joins A k recovers packet k leaves A k Scalable Reliable Multicast Using Multiple Multicast Channels sender loss A orig loss AkAk Kasera, Kurose, Towsley, ACM SIGMETRICS Conference ‘97

Issues how much is performance improved ? receiver, sender processing network bandwidth if (multicast channel IP multicast group), realistically only finite channels available ! overhead of join, leave operations ? router support for multiple multicast channels ?

Analysis rcvrs unicast NAKs to sender infinite channels available system model one sender, R receivers independent loss, probability p NAKs not lost E[Y] = E[Y p ] + pE[Y j ] + pE[Y n ]/(1-p) + p 2 E[Y t ]/(1-p) determined various proc times by instrumenting Linux kernel Y = total per pkt rcv proc time Y p = rcvd pkt proc time Y j = join, leave proc time Y n = NAK proc time Y t = timer proc time NAK processingtimer processing join, leave processing rcvd pkt processing

considerable reduction in rcvr processing costs by using infinite channels example: when R = 1000, p = 0.05, processing cost reduces by approx. 65% similar behavior observed for protocols that multicast NAKs for NAK suppression Receiver Processing Cost Reduction

recycle G retransmission channels, retransmit pkt k on A k mod G example, G = 3 Finite # of Retransmission Channels transmit retransmit pkt 1 on (1) retransmit pkt 4 on (1) lost at r1lost at r2 lost at r1 lost at r2 received at r1 received at r2 received at r

find #unwanted pkts, U, at receiver due to using G channels only model same as before transmit with interval  retransmit with interval  ’ (if pending NAK) U depends upon G, p, R,  /  ’ receiver processing cost, E[Y’] = E[Y] + E[U]E[Y p ] Finite # of Retransmission Channels unwanted pkt processing

How many channels do we need ? find minimum #channels s.t. increase in cost within 1% small #channels for wide range of p,  /  ’, R #channels = 0.5 sensitive to low  /  ’ /’/’

Summary (part 1) use of multiple multicast channels reduces receiver processing small to moderate #channels achieve almost perfect retransmission scoping implementation using router support also saves network bandwidth sender still bottleneck, no improvement in protocol performance

Local Recovery server and/or other receivers aid in loss recovery distribution of loss recovery burden possible reduction in network bandwidth recovery latency retransmission scoping sender loss transmission loss local domains server

scoping retransmissions using multiple multicast channels server-based local recovery performance benefits resource requirements “active” repair services signaling for locating, invoking, revoking services router support summary and future directions Overview

repair servers co-located with routers at strategic locations placement of application level repair service in routers repair servers cache recent pkts receivers, repair servers, recover lost pkts from upper level repair servers, sender repair server Repair Server Based Local Recovery sender receivers Kasera, Kurose, Towsley, IEEE INFOCOM ‘98

Issues how much is performance improved over traditional local recovery approaches ? SRM: dynamically elect receiver for every loss RMTP, LBRM: designated receiver, logger for supplying repairs where to place repair servers ? what are repair server resource requirements ?

based on [YKT ‘97] loss free backbone, sites loss at source link, tails temporally independent loss, probability p sender receivers backbone tail site local domain source link System Model

based on [YKT ‘97] loss free backbone, sites loss at source link, tails temporally independent loss, probability p sender receivers backbone tail site local domain source link System Model designated receiver

based on [YKT ‘97] loss free backbone, sites loss at source link, tails temporally independent loss, probability p sender receivers backbone tail site local domain source link System Model repair server

metrics throughput = 1/max(sender-processing time, receiver- processing time) bandwidth usage = total bytes transmitted over all links per correct transmission analysis: similar approach as in previous problem (optimistic bounds for SRM) Performance Evaluation

repair server-based (RSB) compared to SRM: throughput upto 2.5 times, bandwidth reduction 60% DR-based (DRB): throughput upto 4 times, bandwidth reduction 35% Performance Comparison

additional sender retransmission required if some domains without repair servers place repair servers in high loss domains first homogeneous loss: high % domains require repair server Insufficient Repair Servers 20% tail loss in 20% domains, 1% tail loss in 80% domains

theoretically: infinite realistically: allot finite buffers replace pkts when buffers full if required, replaced pkts recovered upstream size depends upon amount of upstream recovery pkt arrival process, buffer holding time replacement policy Repair Server Buffer Requirements (per session) example: when p = 0.05, 15 buffers ensure almost perfect local recovery

examine three policies FIFO, LRU FIFO-MH: FIFO with minimum buffer holding time = one retransmission interval FIFO-MH shows little improvement over FIFO LRU performs better than FIFO only when #buffers large Buffer Replacement Policies example: arrival rate = 128pkts/sec retransmission interval from round trip time traces

repair server-based approach exhibits superior performance over traditional approaches repair server placement - above loss, higher loss domains first buffer requirement several 10s of buffers (per session) simple FIFO replacement policy sufficient how to make repair server approach dynamic ? Summary (part 2)

scoping retransmissions using multiple multicast channels server-based local recovery performance benefits resource requirements “active” repair services signaling for locating, invoking, revoking services router support summary and future directions Overview

Active Repair Service repair server functionality as active repair service design repair service-based protocol, AER locate, invoke repair services using source path messages (SPMs) minimal router support required for interception of SPM, subcast S SPMRS1 SPMRS2 SPMS SPMs multicast but intercepted NAKs take reverse path

scoping retransmissions using multiple multicast channels server-based local recovery performance benefits resource requirements “active” repair services signaling for locating, invoking, revoking services router support Thesis Contributions

model cost of additional network resources buffer requirements multiple sessions other applications (e.g., web caching) composable multicast services other multicast research revisit IP multicast service model congestion control pricing Future Directions

identify performance enhancing services, examples feedback aggregation selective forwarding repair, rate conversion, log services invoke/revoke services based on application requirements network conditions Composable Multicast Services (Work in Progress) sender protocol feedback aggregation rcvr protocol rcvr protocol rcvr protocol rcvr protocol issues: implementing composability signaling mechanism (SPM++) measurement-based infrastructure rate conversion