Congestion framework for Pseudowires draft-rosen-pwe3-congestion-04.txt Bruce Davie (with Eric Rosen, Stewart Bryant & Luca Martini)

Slides:



Advertisements
Similar presentations
Advanced satellite infrastructures in future global Grid computing: network solutions to compensate delivery delay Blasco Bonito, Alberto Gotta and Raffaello.
Advertisements

© 2004 SafeNet, Inc. All rights reserved. Mobike Protocol Design draft-ietf-mobike-design-00.txt Tero Kivinen
Copyright © 2004 Juniper Networks, Inc. Proprietary and Confidentialwww.juniper.net 1 Point-to-Multipoint Pseudowire Signaling and Auto-Discovery in Layer.
Leading Edge Routing MPLS Enhancements to Support Layer 2 Transport Services Jeremy Brayley
PWE3 Congestion Considerations draft-ietf-pwe3-congcons (temporarily expired) Yaakov (J) Stein David Black Bob Briscoe.
Dynamic Routing Scalable Infrastructure Workshop, AfNOG2008.
Requirements for IP/UDP/RTP header compression To become Editor: Mikael Degermark Input: Charter, 3GPP requirements, contribution from 3G.IP, Editors central.
1 © NOKIA NSIS MIPv6 FW/ November 8 th 2004 Mobile IPv6 - NSIS Interaction for Firewall traversal draft-thiruvengadam-nsis-mip6-fw-01 S. Thiruvengadam.
1 Why Carriers Like Pseudowires… Payload (IP, L2 data, voice) PseudoWires Layer-2 (Ethernet, ATM…) Physical (Optical, Wireless) User Applications Payload.
IP –Based SAN extensions and Performance Thao Pham CS 622 Fall 07.
1 Internet Networking Spring 2003 Tutorial 11 Explicit Congestion Notification (RFC 3168) Limited Transmit (RFC 3042)
A General approach to MPLS Path Protection using Segments Ashish Gupta Ashish Gupta.
1 Internet Networking Spring 2003 Tutorial 11 Explicit Congestion Notification (RFC 3168)
Medium Start in TCP-Friendly Rate Control Protocol CS 217 Class Project Spring 04 Peter Leong & Michael Welch.
A General approach to MPLS Path Protection using Segments Ashish Gupta Ashish Gupta.
MPLS L3 and L2 VPNs Virtual Private Network –Connect sites of a customer over a public infrastructure Requires: –Isolation of traffic Terminology –PE,
TCP/IP Basics A review for firewall configuration.
IETF 59, March 2004Mustapha AïssaouiSlide 1 OAM Procedures for VPWS Interworking draft-aissaoui-l2vpn-vpws-iw-oam-00 Mustapha Aïssaoui, Matthew Bocci,
IP/ICMP Translation Algorithm (IIT) Xing Li, Congxiao Bao, Fred Baker
Extension to LDP-VPLS for Ethernet Broadcast and Multicast draft-delord-l2vpn-ldp-vpls-broadcast-exten-03 Presenter: Zhihua Liu, China Telecom IETF79,
STPP Slide 1 UDP Issues PWE3 – 61 th IETF Yaakov (J) Stein.
Encapsulating MPLS in UDP draft-xu-mpls-in-udp-02 Xiaohu Xu (Huawei) Marshall Eubanks (AmericaFree.TV) Lucy Yong (Huawei) Nischal Sheth.
1 © 2002, Cisco Systems, Inc. All rights reserved. draft-nadeau-pwe3-vccv-00.txt IETF #56 San Francisco, CA USA Thomas D. Nadeau Monique.
Courtesy: Nick McKeown, Stanford 1 TCP Congestion Control Tahir Azim.
{Stewart Bryant, Sami Boutros, Luca Martini,
NECP: the Network Element Control Protocol IETF WREC Working Group November 11, 1999.
Tiziana Ferrari Quality of Service Support in Packet Networks1 Quality of Service Support in Packet Networks Tiziana Ferrari Italian.
Deutsche Telekom Technical Engineering Center. Fat PW Loadbalancing.
EVC Atahar Khan CCIE SP Cisco Systems.
Computer Networks: Multimedia Applications Ivan Marsic Rutgers University Chapter 3 – Multimedia & Real-time Applications.
TDM Pseudowires: “Circuit Breaker” Ideas David Black, EMC March 6, 20141TSVAREA, IETF 89 London.
POSTECH DP&NM Lab. Internet Traffic Monitoring and Analysis: Methods and Applications (1) 4. Active Monitoring Techniques.
PWE3 WG Document Status IETF-62 Danny McPherson Stewart Bryant
Networked & Distributed Systems TCP/IP Transport Layer Protocols UDP and TCP University of Glamorgan.
Datagram Congestion Control Protocol
MaxNet NetLab Presentation Hailey Lam Outline MaxNet as an alternative to TCP Linux implementation of MaxNet Demonstration of fairness, quick.
Internet Protocol ECS 152B Ref: slides by J. Kurose and K. Ross.
1 PWE3 Architecture PWE3 IETF March 2003 Stewart Bryant.
1 Transparent GEHCO Slides for p __luc_gehco-t Lucent Technologies Tom Hiller Pete McCann.
Datagram Congestion Control Protocol (DCCP) CISC TCP/IP and Upper Layer Protocols Presentation by Xiaofeng Han Thanks for Kireeti.
11/27/2015 draft-bocci-bryant-ms-pw-architecture-00.txt An Architecture for Multi-Segment Pseudo Wire Emulation Edge-to-Edge draft-bocci-bryant-pwe3-ms-pw-architecture-00.txt.
MPLS-TP Packet Loss and Delay Measurement draft-frost-mpls-tp-loss-delay-00 Dan Stewart IETF 76 November.
Congestion Issues Stewart Bryant
73rd IETF Minneapolis Nov Framework and Requirements for Virtual Private Multicast Service (VPMS) draft-kamite-l2vpn-vpms-frmwk-requirements-02.txt.
Nov. 8, 2006IDR WG Meeting1 IPv6 Next Hop for IPv4 Prefix In BGP Updates, NH not necessarily of same address family as NLRI Currently deployed examples:
Midterm Review Chapter 1: Introduction Chapter 2: Application Layer
Support for ECN and PCN in MPLS networks draft-davie-ecn-mpls-00.txt Bruce Davie Cisco Systems Bob Briscoe June Tay BT Research.
MPLS over L2TPv3 Encapsulation IETF VersionIHLTOSTotal length IdentificationFlagsFragment offset TTL Protocol ==
1 Advanced Transport Protocol Design Nguyen Multimedia Communications Laboratory March 23, 2005.
CS/EE 145A Reliable Transmission over Unreliable Channel II Netlab.caltech.edu/course.
I E T F 6 3, 3 rd. A U G U S T draft-frost-pwe3-timing-pw-reqs-00 IETF 63 PWE3 Working Group Paris, August 2005.
RMCAT architectural overview Michael Welzl 1 RMCAT, 85 th IETF Meeting
PWE3 Congestion Considerations draft-stein-pwe3-congcons-01.pdf Yaakov (J) Stein David Black Bob Briscoe.
IP1 The Underlying Technologies. What is inside the Internet? Or What are the key underlying technologies that make it work so successfully? –Packet Switching.
IP Pseudowire Florin Balus August, PG 1Florin BalusIETF60 – San Diego Requirements - Existing topology FR/ATM VPNs ATM Network Frame Relay Access.
Uni Innsbruck Informatik th IETF, PMTUD WG: Path MTU Discovery Using Options draft-welzl-pmtud-options-01.txt Michael Welzl
1 Use of PE-PE IP/GRE/IPsec for MPLS PWs draft-raggarwa-pwe3-pw-over-ip- 00.txt Rahul Aggarwal
CS 145A Reliable Communication Netlab.caltech.edu/course.
TCP/IP1 Address Resolution Protocol Internet uses IP address to recognize a computer. But IP address needs to be translated to physical address (NIC).
Lecture 10 Page 1 CS 236 Online Encryption and Network Security Cryptography is widely used to protect networks Relies on encryption algorithms and protocols.
Packet PWE3 – Efficient for IP/MPLS
Introduction There are many situations in which we might use replicated data Let’s look at another, different one And design a system to work well in that.
Quick-Start for TCP and IP
Distributed Systems CS
OAM for Deterministic Networks draft-mirsky-detnet-oam
Editors: Bala’zs Varga, Jouni Korhonen
ECN in QUIC - Questions Surfaced
LOOPS Generic Information Set draft-welzl-loops-gen-info-00
DetNet Data Plane Solutions draft-ietf-detnet-dp-sol-ip-02  draft-ietf-detnet-dp-sol-mpls-02  Bala’zs Varga, Jouni Korhonen, Janos Farkas, Lou Berger,
Distributed Systems CS
Presentation transcript:

Congestion framework for Pseudowires draft-rosen-pwe3-congestion-04.txt Bruce Davie (with Eric Rosen, Stewart Bryant & Luca Martini)

Introduction This is a framework, not a solution draft Tried to examine the issues and list a range of solutions –Tradeoffs to be made in most cases

What is a pseudowire Tunnels (MPLS, GRE, IPSEC, L2TPv3, etc.) Emulated VCs inside tunnels (many-to-one) Attachment VCs (e.g. FR) mapped to emulated VCs PWs are bidirectional (but may only have one-way traffic) dlci 202 Attachment VC Lvc Emulated VC Tunnel Lvc Emulated VC dlci 101 Attachment VC

Why PWs need Congestion Control PWs can carry any sort of traffic, which may not be congestion controlled by the end points Continued health of the Internet requires congestion control of most traffic

Why PWs might not need Congestion Control All the traffic is IP –If UDP, reduce to a previously unsolved problem –If TCP, there’s no need, and prior experience with poor control loop interactions (ATM-ABR) PW service only offered on well-engineered nets, not the Internet at large PW is a premium service that should be able to trample on less important stuff –Never enough PW traffic to congest the net on its own None of these arguments really stand up to scrutiny

Primary Cases of Concern TDM PWs Packet PWs carrying non-congestion- controlled traffic –e.g. Ethernet PW carrying MPEG-2 Deployed over widely shared infrastructure –e.g. the Internet

Design constraints Large number of PWs per edge device –Maintaining TCP-like state per PW considered too costly Hardware data plane implementation typical Existing encapsulations not designed for congestion control Concern about bandwidth efficiency No ACKs in general

Design Choices - Summary How to detect/measure loss rate/congestion? –SEQ numbers or OAM-based or ECN How to feed loss rate/congestion back to sender? –Data, control, or management plane What to do on congestion? –Shape, police, shut-down What to do as congestion abates? Granularity of control –Per-tunnel, per-PW

Detecting Congestion Using sequence numbers has drawbacks –SEQ is optional; using it (today) means misordering becomes loss –False congestion signals if misordering occurs Control/management plane approach –Periodically transmit a count of packets sent in forward direction, count of packets received in reverse direction –Counters in HW, control messages in SW, likely to cause some inaccuracy (as will misordering) –Likely to be less error-prone that SEQ approach –How often - about once per RTT seems needed ECN or PCN –Lack of deployment a concern

Feedback to ingress No reverse data for many PWs Could use a PW control message or an OAM message

TCP Friendly Rate Control RFC3448 –Calculates rate that a TCP connection would get if same loss rate and RTT applied Note: need RTT measurement between PEs Smoother than the standard TCP “sawtooth” Could police/shape the PW or tunnel to that rate Hopefully a no-op Could use local policy to prefer some PWs in a tunnel Could be achieved by selective shutdown of one or more PWs in a tunnel Somewhat tolerant to inaccurate loss measurement Note: already included in FiberChannel PWE spec

Summary of issues Exactly what is the right way to measure congestion & thus set rate? How often to sample loss –Once per RTT seems about right - is less often OK? How to enforce rate? –Shutting off PWs is simple but blunt –Shaping PWs risks TCP interaction –Police-by-dropping considered harmful to TCP Should this be mandatory for all PWs? –Mandatory to implement vs. to enable