PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis.

Slides:



Advertisements
Similar presentations
Data Center Fabrics Lecture 12 Aditya Akella.
Advertisements

With: Radhika Niranjan Mysore, Malveeka Tewari, Ying Zhang (Ericsson Research), Keith Marzullo, Amin Vahdat Meg Walraed-Sullivan University of California,
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
DOT – Distributed OpenFlow Testbed
Introduction into VXLAN Russian IPv6 day June 6 th, 2012 Frank Laforsch Systems Engineer, EMEA
PARIS: ProActive Routing In Scalable Data Centers Dushyant Arora, Theophilus Benson, Jennifer Rexford Princeton University.
J. K. Kim Portland. Trend of Data Center By J. Nicholas Hoover, InformationWeek June 17, :00 AMJ. Nicholas Hoover 200 million Euro Data centers.
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric. Presented by: Vinuthna Nalluri Shiva Srivastava.
Data Center Networking Major Theme: What are new networking issues posed by large-scale data centers? Network Architecture? Topology design? Addressing?
Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, and Amin Vahdat Department.
PortLand: A Scalable Fault- Tolerant Layer 2 Data Center Network Fabric B 財金三 婁瀚升 1.
Data Center Fabrics. Forwarding Today Layer 3 approach: – Assign IP addresses to hosts hierarchically based on their directly connected switch. – Use.
PRESENTED BY: TING WANG PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric Radhika Niranjan Mysore, Andreas Pamboris, Nathan.
1 Version 3.0 Module 8 Virtual LANs. 2 Version 3.0.
Portland: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric Offense Kai Chen Shih-Chi Chen.
PortLand Presented by Muhammad Sadeeq and Ling Su.
Data Center Network Topologies: FatTree
ProActive Routing In Scalable Data Centers with PARIS Joint work with Dushyant Arora + and Jennifer Rexford* + Arista Networks *Princeton University Theophilus.
A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares, Alexander Loukissas, Amin Vahdat Presented by Gregory Peaker and Tyler Maclean.
A Modernize Version of the Antique Telephone Architecture Christopher Francis & Bertha Wilhelm.
A Scalable, Commodity Data Center Network Architecture Mohammad AI-Fares, Alexander Loukissas, Amin Vahdat Presented by Ye Tao Feb 6 th 2013.
A Scalable, Commodity Data Center Network Architecture
A Scalable, Commodity Data Center Network Architecture.
Jennifer Rexford Princeton University MW 11:00am-12:20pm SDN Software Stack COS 597E: Software Defined Networking.
FAR: A Fault-avoidance Routing Method for Data Center Networks with Regular Topology Bin Liu, ZTE.
Cellular Core Network Architecture
Chapter 1: Hierarchical Network Design
Chapter 8 Routing. Introduction Look at: –Routing Basics (8.1) –Address Resolution (8.2) –Routing Protocols (8.3) –Administrative Classification (8.4)
© 2007 Cisco Systems, Inc. All rights reserved.Cisco Public ITE PC v4.0 Chapter 1 1 Connecting to the Network Networking for Home and Small Businesses.
Routing & Architecture
Objectives: Chapter 5: Network/Internet Layer  How Networks are connected Network/Internet Layer Routed Protocols Routing Protocols Autonomous Systems.
Seamless Access to Services for Mobile Users Jennifer Rexford Princeton University Joint work with Matvey Ayre, Mike.
Internetworking – What is internetworking? Connect multiple networks of one or more organizations into a large, uniform communication system. The resulting.
Cloud Scale Performance & Diagnosability Comprehensive SDN Core Infrastructure Enhancements vRSS Remote Live Monitoring NIC Teaming Hyper-V Network.
1 LAN design- Chapter 1 CCNA Exploration Semester 3 Modified by Profs. Ward and Cappellino.
Floodless in SEATTLE : A Scalable Ethernet ArchiTecTure for Large Enterprises. Changhoon Kim, Matthew Caesar and Jenifer Rexford. Princeton University.
Cisco Discovery Protocol. CDP and Router Boot Up When a Cisco device boots up, CDP starts up automatically and allows the device to detect neighbor devices.
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 1: Introduction to Scaling Networks Scaling Networks.
Department of Computer Science A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares Alexander Loukissas Amin Vahdat SIGCOMM’08 Reporter:
© 2015 Mohamed Samir YouTube channel All rights reserved. Samir Agenda Instructor introduction 1. Introduction toEldarin 2.
Portland: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
5: DataLink Layer 5a-1 Bridges and spanning tree protocol Reference: Mainly Peterson-Davie.
Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, Amin Vahdat SIGCOMM.
© 2008 Cisco Systems, Inc. All rights reserved.Cisco ConfidentialPresentation_ID 1 Chapter 1: Hierarchical Network Design Connecting Networks.
15-744: Computer Networking L-12 Data Center Networking I.
Network Virtualization Ben Pfaff Nicira Networks, Inc.
Data Center Networking Major Theme: What are new networking issues posed by large-scale data centers? Network Architecture? Topology design? Addressing?
1 Computer Networks Chapter 5. Network layer The network layer is concerned with getting packets from the source all the way to the destination. Getting.
VL2: A Scalable and Flexible Data Center Network
Data Center Architectures
Data Center Networking
Multi Node Label Routing – A layer 2.5 routing protocol
CIS 700-5: The Design and Implementation of Cloud Networks
Lecture 2: Cloud Computing
Heitor Moraes, Marcos Vieira, Italo Cunha, Dorgival Guedes
Lecture 2: Leaf-Spine and PortLand Networks
Data Center Network Architectures
ETHANE: TAKING CONTROL OF THE ENTERPRISE
Scaling the Network: The Internet Protocol
Chapter 5: Inter-VLAN Routing
NTHU CS5421 Cloud Computing
Aled Edwards, Anna Fischer, Antonio Lain HP Labs
Large-scale (Campus) Lan design (Part I)
Introduction An introduction to the software and organization of the Internet Lab.
NTHU CS5421 Cloud Computing
2018/12/10 Energy Efficient SDN Commodity Switch based Practical Flow Forwarding Method Author: Amer AlGhadhban and Basem Shihada Publisher: 2016 IEEE/IFIP.
VL2: A Scalable and Flexible Data Center Network
ModelNet: A Large-Scale Network Emulator for Wireless Networks Priya Mahadevan, Ken Yocum, and Amin Vahdat Duke University, Goal:
Data Center Architectures
Scaling the Network: The Internet Protocol
Presentation transcript:

PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, and Amin Vahdat SIGCOMM 2009 Presenter: Seokshin Son

Contents Introduction Portland Design Implementation & Evaluation Summary 2/20

INTRODUCTION

Portland in a nutshell PortLand is a single logical layer 2 data center network fabric that scales to millions of endpoints PortLand internally separates host identity from host location –Uses IP address as host identifier –Introduces “Pseudo MAC” (PMAC) addresses internally to encode endpoint location PortLand runs on commodity switch hardware with unmodified hosts 4/20

Goals For Data Center Network Fabrics Easy configuration and management –Plug and play Fault tolerance, routing and addressing –Scalability Commodity switch hardware –Small switch state Virtualization support –Seamless VM migration 5/20

Layer 2 versus Layer 3 Data Center Fabrics TechniquePlug and Play ScalabilitySmall Switch State Seamless VM Migration Layer 2: Flat MAC addresses +--+ Layer 3: IP Addresses /20

PORTLAND DESIGN

PortLand: Plug and Play + Small Switch State PortLand switches learn location in topology using pair-wise communication They assign topologically meaningful addresses to hosts using their location 8/20

PortLand: Main Assumption Hierarchical structure of data center networks: –They are multi-level, multi-rooted trees 9/20

Imposing Hierarchy On A Multi- Rooted Tree PMAC: pod.position.port.vmid 10/20 Pod Position Port Core Aggregation Edge

PortLand: Location Discovery Protocol Location Discovery Messages (LDMs) exchanged between neighboring switches Switches self-discover location on boot up Location characteristicTechnique 1) Tree level / RoleBased on neighbor identity 2) Pod numberAggregation and edge switches agree on pod number 3) Position numberAggregation switches help edge switches choose unique position number 11/20

PortLand: Location Discovery Protocol 12/20

PortLand: Name Resolution 13/20

PortLand: Fabric Manager 14/20

PortLand: Name Resolution 15/20

IMPLEMENTATION & EVAUATION

Portland Prototype 20 OpenFlow NetFPGA switches TCAM + SRAM for flow entries Software MAC rewriting 3 tiered fat-tree 16 end hosts 17/20

PortLand: Evaluation MeasurementsConfigurationResults Network convergence time Keepalive frequency = 10 ms Fault detection time = 50 ms 65 ms TCP convergence timeRTOmin = 200ms~200 ms Multicast convergence time 110ms TCP convergence with VM migration RTOmin = 200ms~200 ms – 600 ms Control traffic to fabric manager 27,000+ hosts, 100 ARPs / sec per host 400 Mbps CPU requirements of fabric manager 27,000+ hosts, 100 ARPs / sec per host 70 CPU cores 18/20

SUMMARY

Summarizing PortLand PortLand is a single logical layer 2 data center network fabric that scales to millions of endpoints Modify network fabric to –Work with arbitrary operating systems and virtual machine monitors –Maintain the boundary between network and end-host administration Scale Layer 2 via network modifications –Unmodified switch hardware and end hosts 20/20