Intra-Domain TE via IGP Metric Tuning Who I Am  Andrew Lange  Exodus, a Cable & Wireless Service  Principal Network Architect 

Slides:



Advertisements
Similar presentations
Responsive Yet Stable Traffic Engineering Srikanth Kandula Dina Katabi, Bruce Davie, and Anna Charny.
Advertisements

Bilal Gonen, Murat Yuksel, Sushil Louis University of Nevada, Reno.
MEASUROUTING: A FRAMEWORK FOR ROUTING ASSISTED TRAFFIC MONITORING PRESENTED BY.
1 Traffic Engineering (TE). 2 Network Congestion Causes of congestion –Lack of network resources –Uneven distribution of traffic caused by current dynamic.
1 EL736 Communications Networks II: Design and Algorithms Class3: Network Design Modeling Yong Liu 09/19/2007.
Bilal Gonen University of Alaska Anchorage Murat Yuksel University of Nevada, Reno.
1 EL736 Communications Networks II: Design and Algorithms Class8: Networks with Shortest-Path Routing Yong Liu 10/31/2007.
Dynamic routing – QoS routing Other approaches to QoS routing Traffic Engineering Practical Traffic Engineering.
Distributed Algorithms for Secure Multipath Routing
Network Architecture for Joint Failure Recovery and Traffic Engineering Martin Suchara in collaboration with: D. Xu, R. Doverspike, D. Johnson and J. Rexford.
Traffic Engineering With Traditional IP Routing Protocols
Towards More Adaptive Internet Routing Mukund Seshadri Prof. Randy Katz.
1 Traffic Engineering for ISP Networks Jennifer Rexford IP Network Management and Performance AT&T Labs - Research; Florham Park, NJ
1IMIC, 8/30/99 Constraint-Based Unicast and Multicast: Practical Issues Bala Rajagopalan NEC C&C Research Labs Princeton, NJ
CSEE W4140 Networking Laboratory Lecture 4: IP Routing (RIP) Jong Yul Kim
More Traffic Engineering TE with MPLS TE in the inter-domain.
Rethinking Internet Traffic Management: From Multiple Decompositions to a Practical Protocol Jiayue He Princeton University Joint work with Martin Suchara,
On Multi-Path Routing Aditya Akella 03/25/02. What is Multi-Path Routing?  Dynamically route traffic Multiple paths to a destination Path taken dependant.
Network Monitoring for Internet Traffic Engineering Jennifer Rexford AT&T Labs – Research Florham Park, NJ 07932
COS 420 Day 17. Agenda Finished Grading Individualized Projects Very large disparity in student grading No two students had same ranking for other students.
Lecture 3. Notations and examples D. Moltchanov, TUT, Spring 2008 D. Moltchanov, TUT, Spring 2015.
Delivery, Forwarding, and Routing
Service Profile-Aware Control Plane: Multi-Instance Fixed Point Approximation within Multi-Granularity VPN Loss Networks Perspective Project Concept Project.
Óscar González de Dios PCE, the magic component of Segment Routing Telefónica I+D.
1 Algorithms for Bandwidth Efficient Multicast Routing in Multi-channel Multi-radio Wireless Mesh Networks Hoang Lan Nguyen and Uyen Trang Nguyen Presenter:
UCSC 1 Aman ShaikhICNP 2003 An Efficient Algorithm for OSPF Subnet Aggregation ICNP 2003 Aman Shaikh Dongmei Wang, Guangzhi Li, Jennifer Yates, Charles.
MATE: MPLS Adaptive Traffic Engineering Anwar Elwalid, et. al. IEEE INFOCOM 2001.
Tomo-gravity Yin ZhangMatthew Roughan Nick DuffieldAlbert Greenberg “A Northern NJ Research Lab” ACM.
TCOM 515 Lecture 6.
Internet Traffic Engineering by Optimizing OSPF Weights Bernard Fortz (Universit é Libre de Bruxelles) Mikkel Thorup (AT&T Labs-Research) Presented by.
L13. Shortest path routing D. Moltchanov, TUT, Spring 2008 D. Moltchanov, TUT, Spring 2014.
Lecture 15. IGP and MPLS D. Moltchanov, TUT, Spring 2008 D. Moltchanov, TUT, Spring 2015.
Shannon Lab 1AT&T – Research Traffic Engineering with Estimated Traffic Matrices Matthew Roughan Mikkel Thorup
The Minimal Communication Cost of Gathering Correlated Data over Sensor Networks EL 736 Final Project Bo Zhang.
1 Chapter-4: Network Flow Modeling & Optimization Deep Medhi and Karthik Ramasamy August © D. Medhi & K. Ramasamy, 2007.
Network Aware Resource Allocation in Distributed Clouds.
Topology aggregation and Multi-constraint QoS routing Presented by Almas Ansari.
Problem Statement The energy and transportation demands continue to rise across the state of Iowa. An investment plan will need to be created for the.
Algorithms for Allocating Wavelength Converters in All-Optical Networks Authors: Goaxi Xiao and Yiu-Wing Leung Presented by: Douglas L. Potts CEG 790 Summer.
Central Control over Distributed Routing fibbing.net SIGCOMM Stefano Vissicchio 18th August 2015 UCLouvain Joint work with O. Tilmans (UCLouvain), L. Vanbever.
Presenter: Jonathan Murphy On Adaptive Routing in Wavelength-Routed Networks Authors: Ching-Fang Hsu Te-Lung Liu Nen-Fu Huang.
Minimax Open Shortest Path First (OSPF) Routing Algorithms in Networks Supporting the SMDS Service Frank Yeong-Sung Lin ( 林永松 ) Information Management.
TCOM 509 – Internet Protocols (TCP/IP) Lecture 06_a Routing Protocols: RIP, OSPF, BGP Instructor: Dr. Li-Chuan Chen Date: 10/06/2003 Based in part upon.
Intradomain Traffic Engineering By Behzad Akbari These slides are based in part upon slides of J. Rexford (Princeton university)
6 December On Selfish Routing in Internet-like Environments paper by Lili Qiu, Yang Richard Yang, Yin Zhang, Scott Shenker presentation by Ed Spitznagel.
On Selfish Routing In Internet-like Environments Lili Qiu (Microsoft Research) Yang Richard Yang (Yale University) Yin Zhang (AT&T Labs – Research) Scott.
An Approach to IP Network Traffic Engineering NANOG Miami, FL Chris Liljenstolpe Cable & Wireless
1 An Arc-Path Model for OSPF Weight Setting Problem Dr.Jeffery Kennington Anusha Madhavan.
1 Slides by Yong Liu 1, Deep Medhi 2, and Michał Pióro 3 1 Polytechnic University, New York, USA 2 University of Missouri-Kansas City, USA 3 Warsaw University.
© 2005 Cisco Systems, Inc. All rights reserved. BGP v3.2—6-1 Scaling Service Provider Networks Scaling IGP and BGP in Service Provider Networks.
HELSINKI UNIVERSITY OF TECHNOLOGY Visa Holopainen 1/18.
Spring 2000CS 4611 Routing Outline Algorithms Scalability.
1 Traffic Engineering By Kavitha Ganapa. 2 Introduction Traffic engineering is concerned with the issue of performance evaluation and optimization of.
Internet Traffic Engineering Motivation: –The Fish problem, congested links. –Two properties of IP routing Destination based Local optimization TE: optimizing.
System & Network Reading Group On Selfish Routing In Internet-Like Evironments Lili Qiu (Microsoft Research) Yang Richard Yang (Yale University) Yin Zhang.
1 Netflow Collection and Aggregation in the AT&T Common Backbone Carsten Lund.
Interaction and Animation on Geolocalization Based Network Topology by Engin Arslan.
Presented by Tae-Seok Kim
ECE 544: Traffic engineering (supplement)
Routing.
An Equal-Opportunity-Loss MPLS-Based Network Design Model
(How the routers’ tables are filled in)
Frank Yeong-Sung Lin (林永松) Information Management Department
Computer Simulation of Networks
Netscope: Traffic Engineering for IP Networks
Frank Yeong-Sung Lin (林永松) Information Management Department
Advisor: Yeong-Sung, Lin, Ph.D. Presented by Yu-Ren, Hsieh
Computer Networks Protocols
Routing.
2019/9/14 The Deep Learning Vision for Heterogeneous Network Traffic Control Proposal, Challenges, and Future Perspective Author: Nei Kato, Zubair Md.
Presentation transcript:

Intra-Domain TE via IGP Metric Tuning

Who I Am  Andrew Lange  Exodus, a Cable & Wireless Service  Principal Network Architect   Successfully navigated the straights of Chapter 11, between Scylla & Charybdis...and somehow ended up in Britain.

What this IS  This IS an introduction to the wonderful world of using flow information to tune your IGP metrics.

What this is NOT  This is NOT an end-all be-all guide to how to optimize your IGP metrics.

Problem We're Trying to Solve  How can we maximally* utilize our network infrastructure, without adding the complications associated with MPLS?  Can this even be done?  Well, of course it can, or this presentation would be remarkably short.  Why ?  The more we can get out of our network the more cost effective it becomes, and the happier the finance people get. Plus, it's cool.

*Maximally  Maximally is a thorny term. Long story short:  Optimum network flows can be represented as shortest paths with respect to a set of positive link weights (Wang, Wang & Zhang).  With current IGP's, determining optimum is NP-Hard, BUT, very close (within 3%) approximations can be made (Fortz & Thorup).

Scholarship Abounds  The first concrete way of doing this that we ran across was in Fortz & Thorup's paper Internet Traffic Engineering by Optimizing OSPF Weights.  This literature tends to be quite recent (1999 and newer).  How easy it is to determine the optimal values very much depends on what your network and flows look like.

What is Required  Accurate flow data between each set of backbone nodes.  An optimization routine to apply to the flow data.

Getting the Flow Data

Getting the Data - Tool Selection  Using Netflow is the only way to gather a traffic matrix without using an overlay design.  Looked at a variety of options, including building our own, and settled on the Ixia (nee Caimis) product.

Getting the Data - Our Issues  Vendor HW/SW combinations are not always supported with the netflow feature set.  Full deployment of the tools is pending operational deployment of the right code base.  Sampling rate needs to be grossed up.

Getting the Data - Configuration  How have we configured our collectors?  Data is collected on the interfaces inbound into the backbone routers from the datacenter.  Flow data is sampled at 1:100.  Collectors peer with the backbone routers as route-reflector clients.  Collectors gather, among other things, BGP NextHop information.

Node Overview

Munging the Data- Basics  How do we process the collected data?  Data is summarized daily.  To assemble the flow matrix, data is aggregated across the interfaces and the routers for a given site. There are some problems with this.

Munging the Data - Problems  Data is an average, peak utilization is not available. This is probably ok for this application, since average and peak tend to follow the same proportions. But we're working on getting peak to compare the results using that data.  Assumes both routers function as one (Nodal Aggregation). This is useful to simplify things as we first work out the models, but we will need to get more detailed as the models are refined.

Munging the Data - Aggregating the Flows  Aggregated daily summaries are post processed with a script that correlates BGP NextHop with destination datacenter and combines the flows destined for that datacenter.  Currently does not gross up flow size to compensate for sampling.

Flow Matrix - Example (sntc08)

Building a Model

Offline vs. Online  We have chosen to pursue offline metric optimization.  Online, or dynamic, metric optimization imposes a whole other set of requirements, such as speed of the optimization model, and lot of trust.  At least at this point, our target for intra- domain TE is in the medium/long term timeframe. If we are running our network so hot that we have to reoptimize multiple times a day, then we need more bandwidth.

Modeling Assumptions  Model assumes that when flows grow the proportions remain the same.  Model does not take flow splitting (ECMP) into account currently. Except ECMP between two adjacent nodes, which is represented by increasing the size of the link between them in the model.

What follows is an Example  10 nodes, 15 links (30 arcs).  10 demand sets.  Real backbone network would be more complicated, but findings still hold.

Example Data  Because we were not able to poll the full matrix of data from the network, the data we're using for this example is extrapolated from the flow data we do have. It is only approximate.

Example Network Diagram

Example Network Info  All links are OC48.  There are no nodal constraints (i.e. Routers are assumed to be able to push line rate.)

Base Demands

Current Metrics  Current Metrics are agnostic to flow information (based on RTT between nodes).  Under current loads the example network is nicely overprovisioned.  We're going to focus on how much more load we can put on this network before any link exceeds 80% utilization (to allow for microbursts, etc.) This is 1990 Mbps for an OC48. We are going to do this by increasing the values of the Base Demands.

Shortest Paths - Current Metrics

Link Loading - Current Metrics

Abracadabra!  Sample run of one of the models:  ampl: model fixtwo-int.mod; data cap- 3.3.dat; solve;  CPLEX 7.1.0: optimal integer solution; objective  31 MIP simplex iterations  0 branch-and-bound nodes

A Bit Behind the Curtain  Using AMPL/CPLEX to define the models.  This consists of a model file specifying the model's:  Objective (e.g. Minimize overall network load).  Constraints (e.g. Do not exceed capacity on links.)  And a data file, which specifies:  What the network looks like.  What the demands are.

Shortest Paths - New Metrics

Link Loading - New Metrics

Resources and Thanks

Optimization Resources - Papers  Sample Papers:  Internet Traffic Engineering by Optimizing OSPF Weights, Fortz & Thorup.  Internet Traffic Engineering without Full Mesh Overlaying, Wang, Wang & Zhang.  Dynamic Optimization of OSPF Weights using Online Simulation, Ye, et. al.

Optimization Resources - Tools  Mathematical Programming Tools  AMPL/CPLEX (  OPL/CPLEX ( )

Special Thanks To  Dr. Irv Lustig, for invaluable help with the modeling languages.