ACN: CSFQ1 CSFQ Core-Stateless Fair Queueing Presented by Nagaraj Shirali Choong-Soo Lee ACN: CSFQ1.

Slides:



Advertisements
Similar presentations
Using Edge-To-Edge Feedback Control to Make Assured Service More Assured in DiffServ Networks K.R.R.Kumar, A.L.Ananda, Lillykutty Jacob Centre for Internet.
Advertisements

RED-PD: RED with Preferential Dropping Ratul Mahajan Sally Floyd David Wetherall.
RED Enhancement Algorithms By Alina Naimark. Presented Approaches Flow Random Early Drop - FRED By Dong Lin and Robert Morris Sabilized Random Early Drop.
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.
CS 268: Lecture 8 Router Support for Congestion Control Ion Stoica Computer Science Division Department of Electrical Engineering and Computer Sciences.
CS 4700 / CS 5700 Network Fundamentals Lecture 12: Router-Aided Congestion Control (Drop it like it’s hot) Revised 3/18/13.
Network Border Patrol Celio Albuquerque, Brett J. Vickers and Tatsuya Suda Jaideep Vaidya CS590F Fall 2000.
Network Border Patrol: Preventing Congestion Collapse and Promoting Fairness in the Internet Celio Albuquerque, Brett J. Vickers, Tatsuya Suda 1.
The War Between Mice and Elephants LIANG GUO, IBRAHIM MATTA Computer Science Department Boston University ICNP (International Conference on Network Protocols)
Ion Stoica, Scott Shenker, and Hui Zhang SIGCOMM’98, Vancouver, August 1998 subsequently IEEE/ACM Transactions on Networking 11(1), 2003, pp Presented.
Max Min Fairness How define fairness? “ Any session is entitled to as much network use as is any other ” ….unless some sessions can use more without hurting.
XCP: Congestion Control for High Bandwidth-Delay Product Network Dina Katabi, Mark Handley and Charlie Rohrs Presented by Ao-Jan Su.
1 Core Stateless Fair Queueing Ion Stoica Hui Zhang Scott Shenker CMU CMU Xerox PARC CMU CMU Xerox PARC.
Network Congestion Gabriel Nell UC Berkeley. Outline Background: what is congestion? Congestion control – End-to-end – Router-based Economic insights.
1 Core-Stateless Fair Queueing: A Scalable Architecture to Approximate Fair Bandwidth Allocations in High Speed Networks Core-Stateless Fair Queueing:
The War Between Mice and Elephants Presented By Eric Wang Liang Guo and Ibrahim Matta Boston University ICNP
1 Minseok Kwon and Sonia Fahmy Department of Computer Sciences Purdue University {kwonm, All our slides and papers.
Networking Issues in LAN Telephony Brian Yang
Low Delay Marking for TCP in Wireless Ad Hoc Networks Choong-Soo Lee, Mingzhe Li Emmanuel Agu, Mark Claypool, Robert Kinicki Worcester Polytechnic Institute.
Achieving End-to-End Fairness in Wireless Networks Ananth Rao Ion Stoica OASIS Retreat, Jul 2005.
Katz, Stoica F04 EECS 122: Introduction to Computer Networks Packet Scheduling and QoS Computer Science Division Department of Electrical Engineering and.
CS 268: Lecture 8 (Router Support for Congestion Control) Ion Stoica February 19, 2002.
1 Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks Ion Stoica,Scott Shenker, and Hui Zhang SIGCOMM’99,
A Real-Time Video Multicast Architecture for Assured Forwarding Services Ashraf Matrawy, Ioannis Lambadaris IEEE TRANSACTIONS ON MULTIMEDIA, AUGUST 2005.
ACN: TCP Friendly1 Promoting the Use of End-to-End Congestion Control in the Internet Sally Floyd and Kevin Fall IEEE/ACM Transactions on Networking May.
1 Emulating AQM from End Hosts Presenters: Syed Zaidi Ivor Rodrigues.
Active Queue Management Rong Pan Cisco System EE384y Spring Quarter 2006.
Random Early Detection Gateways for Congestion Avoidance
The War Between Mice and Elephants By Liang Guo (Graduate Student) Ibrahim Matta (Professor) Boston University ICNP’2001 Presented By Preeti Phadnis.
1 Core-Stateless Fair Queueing: Achieving Approximately Fair Bandwidth Allocations in High Speed Networks Ion Stoica,Scott Shenker, and Hui Zhang SIGCOMM’99,
Core Stateless Fair Queueing Stoica, Shanker and Zhang - SIGCOMM 98 Rigorous fair Queueing requires per flow state: too costly in high speed core routers.
CS 268: Lecture 17 (Dynamic Packet State) Ion Stoica April 15, 2002.
Lightweight Active Router-Queue Management for Multimedia Networking M. Parris, K. Jeffay, and F.D. Smith Department of Computer Science University of.
UCB Improvements in Core-Stateless Fair Queueing (CSFQ) Ling Huang U.C. Berkeley cml.me.berkeley.edu/~hlion.
Advanced Computer Networks : RED 1 Random Early Detection Gateways for Congestion Avoidance Sally Floyd and Van Jacobson, IEEE Transactions on Networking,
Core Stateless Fair Queueing Stoica, Shanker and Zhang - SIGCOMM 98 Fair Queueing requires per flow state: too costly in high speed core routers Yet, some.
Packet Scheduling From Ion Stoica. 2 Packet Scheduling  Decide when and what packet to send on output link -Usually implemented at output interface 1.
Advance Computer Networking L-5 TCP & Routers Acknowledgments: Lecture slides are from the graduate level Computer Networks course thought by Srinivasan.
Advance Computer Networking L-6 TCP & Routers Acknowledgments: Lecture slides are from the graduate level Computer Networks course thought by Srinivasan.
ACN: RED paper1 Random Early Detection Gateways for Congestion Avoidance Sally Floyd and Van Jacobson, IEEE Transactions on Networking, Vol.1, No. 4, (Aug.
Wolfgang EffelsbergUniversity of Mannheim1 Differentiated Services for the Internet Wolfgang Effelsberg University of Mannheim September 2001.
1 On Class-based Isolation of UDP, Short-lived and Long-lived TCP Flows by Selma Yilmaz Ibrahim Matta Computer Science Department Boston University.
Fair Queueing. 2 First-Come-First Served (FIFO) Packets are transmitted in the order of their arrival Advantage: –Very simple to implement Disadvantage:
Worcester Polytechnic Insitute, Worcester, MA, USA1 Traffic Sensitive Active Queue Management for Improved Multimedia Streaming Authors: Vishal Phirke,
Presented by: Peng Wang EE Department University of Delaware A Probabilistic Approach for Achieving Fair Bandwidth Allocation in CSFQ.
Queueing and Active Queue Management Aditya Akella 02/26/2007.
9.7 Other Congestion Related Issues Outline Queuing Discipline Avoiding Congestion.
Packet Scheduling and Buffer Management Switches S.Keshav: “ An Engineering Approach to Networking”
EE 122: Lecture 15 (Quality of Service) Ion Stoica October 25, 2001.
T. S. Eugene Ngeugeneng at cs.rice.edu Rice University1 COMP/ELEC 429 Introduction to Computer Networks Lecture 18: Quality of Service Slides used with.
We used ns-2 network simulator [5] to evaluate RED-DT and compare its performance to RED [1], FRED [2], LQD [3], and CHOKe [4]. All simulation scenarios.
1 Core-Stateless Fair Queueing: A Scalable Architecture to Approximate Fair Bandwidth Allocations in High Speed Networks Core-Stateless Fair Queueing:
XCP: eXplicit Control Protocol Dina Katabi MIT Lab for Computer Science
Spring Computer Networks1 Congestion Control Sections 6.1 – 6.4 Outline Preliminaries Queuing Discipline Reacting to Congestion Avoiding Congestion.
ECEN 619, Internet Protocols and Modeling Prof. Xi Zhang Random Early Detection Gateways for Congestion Avoidance Sally Floyd and Van Jacobson, IEEE Transactions.
Univ. of TehranIntroduction to Computer Network1 An Introduction Computer Networks An Introduction to Computer Networks University of Tehran Dept. of EE.
1 Flow-Aware Networking Introduction Concepts, graphics, etc. from Guide to Flow-Aware Networking: Quality-of-Service Architectures and Techniques for.
Internet Quality of Service
Internet Quality of Service
Corelite Architecture: Achieving Rated Weight Fairness
Congestion Control and Resource Allocation
Core-Stateless Fair Queueing: A Scalable Architecture to Approximate Fair Bandwidth Allocations in High Speed Networks Ion Stoica, Scott Shenker, and Hui.
EE 122: Router Support for Congestion Control: RED and Fair Queueing
Random Early Detection Gateways for Congestion Avoidance
Advance Computer Networking
Max Min Fairness How define fairness?
Advance Computer Networking
COMP/ELEC 429 Introduction to Computer Networks
Congestion Control and Resource Allocation
Presentation transcript:

ACN: CSFQ1 CSFQ Core-Stateless Fair Queueing Presented by Nagaraj Shirali Choong-Soo Lee ACN: CSFQ1

2 About the Authors Ion Stoica – CMU PhD degree from Carnegie Mellon University Assistant Professor at University of California, Berkeley Networking with an emphasis on Quality of Service and traffic management in the Internet Hui Zhang – CMU PhD degree from University of California, Berkeley Associate Professor at Carnegie Mellon University Internet, multimedia systems, resource management, and performance analysis Scott Shenker – Xerox PARC Chair for the Integrated Services (INTSERV) charter ACN: CSFQ2

3 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ4 Introduction Main Idea: - Achieve fair bandwidth allocations at the router without the implementation complexity usually associated with it. Goals: - Achieve fair allocation close to Fair Queueing and comparable or better than RED and FRED under most scenarios. - Reduce complexity by not having the core node maintain per flow state. - Approximate weighted FQ.

ACN: CSFQ5 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ6 Previous Work FIFO queueing with Drop Tail Random Early Drop (RED) Flow Random Early Drop (FRED) Fair Queueing (FQ)

ACN: CSFQ7 FIFO queueing with Drop Tail SERVERFIFO Disadvantages: Pushes congestion control out to end hosts (TCP) Introduces global synchronization when packets are dropped from several connections

ACN: CSFQ8 Random Early Drop (RED) SERVERFIFO MinthMaxth MinthMaxth P(drop) 1. 0 MaxP Disadvantage: For web traffic, RED provides no clear advantage over tail-drop FIFO for end- user response times

ACN: CSFQ9 Flow Random Early Drop (FRED) SERVERFIFO Disadvantage: Complex to implement – maintain state on per-flow basis

ACN: CSFQ10 Fair Queueing Disadvantage: Need to perform packet classification and maintain state and buffers on per-flow basis and perform operations on per-flow basis

ACN: CSFQ11 Definitions Island of routers – a contiguous portion of the network with well defined interior and edges. Edge Router – computes per-flow rate estimates and labels the packets with these estimates. Core Router – uses FIFO queueing and keeps no per- flow state, employs a probabilistic dropping algorithm that uses the packet label and its own measurement of aggregate traffic. Stateless – absence of per-flow state at the core routers.

ACN: CSFQ12 Island of Routers Source: CSFQ, Stoica, Berkeley

ACN: CSFQ13 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ14 CSFQ and its Algorithms Assumptions: Fair Allocation methods like FQ are necessary for congestion control. The complexity involved is a major hindrance to their adoption.

ACN: CSFQ15 CSFQ In an island of routers, edge routers compute per-flow rate estimates and label the packets with these estimates. Core routers use FIFO queueing and keep no per-flow state, they employ a probabilistic dropping algorithm based on packet labels and own aggregate traffic estimates.

ACN: CSFQ16 CSFQ Bandwidth allocations using this method are approximately fair. Core routers keep no per-flow state and avoid using complicated packet scheduling and buffering algorithms, hence are easier to adopt.

ACN: CSFQ17 Assume that flow i has arrival rate r i (t) and the fair rate is a(t). If r i (t) < a(t), all of its traffic is forwarded. If r i (t) > a(t), then a fraction (r i (t) - a(t))/ r i (t) will be dropped; each packet of the flow is dropped with probability (1-a(t)/r i (t)). Thus the output rate of any flow i will be max(r i (t),a(t)). CSFQ

ACN: CSFQ18 CSFQ The problem now becomes how to calculate the flow rate r i (t) values and the fair rate a(t), without keeping per flow state in the core routers. Flow rates r i (t), are calculated at edge routers which keep per flow state and then insert the rate value inside the packet header of packets belonging to that flow.

ACN: CSFQ19 CSFQ To estimate the fair rate a(t), an iterative procedure is used: core routers estimate aggregate arrival rate A and the aggregate rate of accepted traffic F (arrival rate – dropped packets). Based on these, the fair rate a is computed periodically as: - if there is no congestion (A<=C where C is the link’s capacity), then a is set to the maximum r i (t) - if the links are congested, then a new = a old *C/F

ACN: CSFQ20 CSFQ - Example Assume we have two flows f 1 and f 2, with rates r 1 = 20 and r 2 = 30 and the link’s capacity is C = 30. Initially let’s say that only r 1 is active and the link is not congested, so a 1 = 20. Then r 2 becomes active. Since no packets were dropped, F = 50. Since A = 50>C, a 2 = a 1 * C/F = 20 * 30/50 = 12 Therefore, for f 1 (1-12/20 = 40%) of its packets are dropped while for f 2 (1-12/30 = 60%) of its packets are dropped and F = = 24 Since A>C, a 3 = a 2 * C/F = 12 * 30/24 = 15 Now F = 30, and a 4 = a 3 * C/F = 15 * 30/30 = 15. Therefore, a has converged to the right fair rate. Source: Network Reading Group, Stoica

ACN: CSFQ21 CSFQ Estimation of flow arrival rates: R new = (1-e -T/K )*l/T + e -T/K *R old where T = packet interarrival time l = packet size K = constant To summarize, Edge routers needs to 1)Classify the packet to a flow 2)Update the fair share rate estimation for the outgoing link 3)Update the flow rate estimation 4)Label the packet

ACN: CSFQ22 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ23 Simulations – Single Congested Link Mbps UDP Flows

ACN: CSFQ24 Simulations – Single Congested Link

ACN: CSFQ25 Simulations – Single Congested Link Mbps TCP Flows UDP Flow UDP flows at 10Mbps 10Mbps

ACN: CSFQ26 Simulations – Single Congested Link

ACN: CSFQ27 Simulations – Single Congested Link N 10Mbps UDP Flows TCP Flow

ACN: CSFQ28 Simulations – Single Congested Link

ACN: CSFQ29 Simulations – Multiple Congested Links TCP/UDP Source TCP/UDP Sink UDP1UDP10 Sources Sinks UDP1UDP10 10Mbps Links

ACN: CSFQ30 Simulations – Multiple Congested Links UDP

ACN: CSFQ31 Simulations – Multiple Congested Links TCP

ACN: CSFQ32 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes RLM (Receiver-driven Layered Multicast) Only first 5 layers (~0.992Mbps) TCP-friendly like 3 RLM flows and 1 TCP flow

ACN: CSFQ33 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes FIFO

ACN: CSFQ34 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes RED

ACN: CSFQ35 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes FRED

ACN: CSFQ36 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes DRR

ACN: CSFQ37 Simulations – Coexistence of Adaptation Schemes Simulations – Coexistence of Adaptation Schemes CSFQ

ACN: CSFQ38 Simulations – Different Traffic Models 1 On/Off Flows 100ms on, 1900ms off Rate : 10Mbps Sends 6758 packets 19 competing TCP flows

ACN: CSFQ39 Simulations – Different Traffic Models AlgorithmDeliveredDropped DRR CSFQ FRED RED FIFO

ACN: CSFQ40 Simulations – Different Traffic Models 60 TCP Flows Exponentially distributed inter-arrival times with mean of 0.05ms Pareto distributed transfer time with mean of 20 packets 1 UDP flow (10Mbps)

ACN: CSFQ41 Simulations – Different Traffic Models AlgorithmMean timeStd. dev DRR2599 CSFQ62142 FRED40174 RED FIFO

ACN: CSFQ42 Simulations – Large Latency 10Mbps link with 100ms latency 1 UDP flow at 10Mbps 19 TCP flows AlgorithmMeanStd. dev DRR CSFQ FRED RED62880 FIFO37869

ACN: CSFQ43 Simulations – Packet Relabeling 10Mbps links Sink Sources

ACN: CSFQ44 Simulations – Packet Relabeling TrafficFlow 1Flow 2Flow 3 UDP TCP

ACN: CSFQ45 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ46 Evaluations of CSFQ Reasonable approximation of fair share Roughly comparable performance to FRED Sometimes much better than FRED Note : FRED has per-packet overhead Not quite as fair as DRR

ACN: CSFQ47 Outline Introduction Background: Definitions and Previous Work CSFQ and its Algorithms Simulations Evaluations of CSFQ Conclusions and Future Work

ACN: CSFQ48 Conclusions and Future Work CSFQ rate-based active queue management Rate estimation at the edge and packet labels for core routers Large latency effect Possible extension of CSFQ for QoS

ACN: CSFQ49 Back-up Slide(s) Slide 2 -Ion Stoica – research interest is to develop techniques and architectures that allow powerful and flexible network services to be deployed in the Internet without compromising its scalability and robustness. -Scott Shenker - The working group will focus on defining a minimal set of global requirements which transition the Internet into a robust integrated-service communications infrastructure. Slide 4 - Congestion today (1998) is controlled by end-hosts (TCP) -FQ – has to maintain state, manage buffers, perform packet scheduling on per-flow basis. Slide 8 -SFloyd, Jacobson, 93. For long-lived TCP connections like file transfer, it might make a difference. Slide 9 -Dong Lin, Robert Morris in 1997 – works well with different traffic – TCP and UDP etc. Slide 10 -DDR – Deficit Round Robin or WFQ. Slide 21 -Exponential average to estimate the rate of flow since this closely reflects a fluid averaging process which is independent of the packetizing structure. And the solution is bounded as it converges to a real value.