Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, and Amin Vahdat Department.

Slides:



Advertisements
Similar presentations
EdgeNet2006 Summit1 Virtual LAN as A Network Control Mechanism Tzi-cker Chiueh Computer Science Department Stony Brook University.
Advertisements

Data Center Fabrics Lecture 12 Aditya Akella.
CCNA3 v3 Module 7 v3 CCNA 3 Module 7 JEOPARDY K. Martin.
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
Computer Networks21-1 Chapter 21. Network Layer: Address Mapping, Error Reporting, and Multicasting 21.1 Address Mapping 21.2 ICMP 21.3 IGMP 21.4 ICMPv6.
Packet Switching COM1337/3501 Textbook: Computer Networks: A Systems Approach, L. Peterson, B. Davie, Morgan Kaufmann Chapter 3.
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.
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.
Switching Topic 4 Inter-VLAN routing. Agenda Routing process Routing VLANs – Traditional model – Router-on-a-stick – Multilayer switches EtherChannel.
Week 5: Internet Protocol Continue to discuss Ethernet and ARP –MTU –Ethernet and ARP packet format IP: Internet Protocol –Datagram format –IPv4 addressing.
1 Version 3.0 Module 8 Virtual LANs. 2 Version 3.0.
Subnetting.
1 Network Layer: Host-to-Host Communication. 2 Network Layer: Motivation Can we built a global network such as Internet by extending LAN segments using.
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.
COS 461: Computer Networks
A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares, Alexander Loukissas, Amin Vahdat Presented by Gregory Peaker and Tyler Maclean.
A Study of MPLS Department of Computing Science & Engineering DE MONTFORT UNIVERSITY, LEICESTER, U.K. By PARMINDER SINGH KANG
A Scalable, Commodity Data Center Network Architecture.
Spanning Tree and Multicast. The Story So Far Switched ethernet is good – Besides switching needed to join even multiple classical ethernet networks Routing.
MPLS And The Data Center Adrian Farrel Old Dog Consulting / Juniper Networks
FAR: A Fault-avoidance Routing Method for Data Center Networks with Regular Topology Bin Liu, ZTE.
Protocol Headers Pre DA SA 0800h … version H L 6 TCP Header Data FCS
ICMP (Internet Control Message Protocol) Computer Networks By: Saeedeh Zahmatkesh spring.
Network Redundancy Multiple paths may exist between systems. Redundancy is not a requirement of a packet switching network. Redundancy was part of the.
23-Support Protocols and Technologies Dr. John P. Abraham Professor UTPA.
Chapter 22 Network Layer: Delivery, Forwarding, and Routing
© Janice Regan, CMPT 128, CMPT 371 Data Communications and Networking Multicast routing.
Router and Routing Basics
CMPT 471 Networking II Address Resolution IPv4 ARP RARP 1© Janice Regan, 2012.
NUS.SOC.CS2105 Ooi Wei Tsang Application Transport Network Link Physical you are here.
Fall 2005Computer Networks20-1 Chapter 20. Network Layer Protocols: ARP, IPv4, ICMPv4, IPv6, and ICMPv ARP 20.2 IP 20.3 ICMP 20.4 IPv6.
Floodless in SEATTLE : A Scalable Ethernet ArchiTecTure for Large Enterprises. Changhoon Kim, Matthew Caesar and Jenifer Rexford. Princeton University.
CS 447 Networks and Data Communication ARP (Address Resolution Protocol) for the Internet Department of Computer Science Southern Illinois University Edwardsville.
Campus Networking Best Practices Hervey Allen NSRC & University of Oregon Dale Smith University of Oregon & NSRC
Department of Computer Science A Scalable, Commodity Data Center Network Architecture Mohammad Al-Fares Alexander Loukissas Amin Vahdat SIGCOMM’08 Reporter:
Transport Layer3-1 Chapter 4: Network Layer r 4. 1 Introduction r 4.2 Virtual circuit and datagram networks r 4.3 What’s inside a router r 4.4 IP: Internet.
Routing protocols. Static Routing Routes to destinations are set up manually Route may be up or down but static routes will remain in the routing tables.
Internet Protocols. ICMP ICMP – Internet Control Message Protocol Each ICMP message is encapsulated in an IP packet – Treated like any other datagram,
1 © 2003, Cisco Systems, Inc. All rights reserved. CCNA 3 v3.0 Module 7 Spanning Tree Protocol.
Portland: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric
By: Muhammad Hanif.  Have a heart that never harden, and a temper that never tire, and a touch that never hurt.  The True happiness is to give love.
CS470 Computer Networking Protocols Huiping Guo Department of Computer Science California State University, Los Angeles 4. Internetworking.
1 © 2004, Cisco Systems, Inc. All rights reserved. CCNA 1 Module 10 Routing Fundamentals and Subnets.
Cisco Confidential © 2013 Cisco and/or its affiliates. All rights reserved. 1 Cisco Networking Training (CCENT/CCT/CCNA R&S) Rick Rowe Ron Giannetti.
Ad Hoc On-Demand Distance Vector Routing (AODV) ietf
Inter-domain Routing Outline Border Gateway Protocol.
Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, Amin Vahdat SIGCOMM.
ADDRESS MAPPING ADDRESS MAPPING The delivery of a packet to a host or a router requires two levels of addressing: logical and physical. We need to be able.
PortLand: A Scalable Fault-Tolerant Layer 2 Data Center Network Fabric Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis.
Data Center Architectures
NAT – Network Address Translation
Scaling the Network Chapters 3-4 Part 2
Heitor Moraes, Marcos Vieira, Italo Cunha, Dorgival Guedes
Lecture 2: Leaf-Spine and PortLand Networks
ETHANE: TAKING CONTROL OF THE ENTERPRISE
Scaling the Network: The Internet Protocol
Revisiting Ethernet: Plug-and-play made scalable and efficient
ICMP ICMP – Internet Control Message Protocol
: An Introduction to Computer Networks
NTHU CS5421 Cloud Computing
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
Data Center Architectures
Scaling the Network: The Internet Protocol
IS-IS VPLS for Data Center Network draft-xu-l2vpn-vpls-isis-02
Presentation transcript:

Radhika Niranjan Mysore, Andreas Pamboris, Nathan Farrington, Nelson Huang, Pardis Miri, Sivasankar Radhakrishnan, Vikram Subramanya, and Amin Vahdat Department of Computer Science and Engineering University of California San Diego

 To find out suitable topology for datacenter network.  To design new address and protocol to satisfy the requirement of plug-and-play.  To build a fabric on layer 2 that gets the scalability, fault tolerance, easy management and flexibility  To implement the new fabric and evaluate it.

 Large scale: Thousands of servers  Virtualization: VM in end host  Substantial portion of Internet communication will take place within data center networks  Cost concerned

 Any VM may migrate to any physical machine. Migrating VMs should not have to change their IP addresses as doing so will break pre-existing TCP connections and application- level state.  An administrator should not need to configure any switch before deployment.  Any end host should be able to efficiently communicate with any other end host in the data center along any of the available physical communication paths.  There should be no forwarding loops.  Failures will be common at scale, so failure detection should be rapid and efficient. Existing unicast and multicast sessions should proceed unaffected to the extent allowed by underlying physical connectivity.

 ISIS can detect failure  Less administrative  Agnostic to the IP address of a VM.  Employing a single spanning tree to avoid forwarding loops  Bad scalability(broadcast)  Topology problem  VLAN still can’t deal with scalability  OSPF can detect failure  TTL can help to reduce forwarding loops  Administrative burden  Difficult to diagnose unreachable hosts  Moving VM will lead to re- allocate IP address

 Traditional  Fat Tree

PodPositionPortVmid 16Bit 8Bit

 Proxy-based: rely on fabric manager to reduce broadcast overhead Step 1: An edge switch intercepts an ARP request for an IP to MAC address mapping and forwards the request to the fabric manager. Step 2: The fabric manager consults its PMAC table to see if an entry is available for the target IP address. Step 3: Returns the PMAC to the edge switch. Step 4: The edge switch creates an ARP reply and returns it to the original host.

 When no mapping: Fall back to broadcast to all end hosts to retrieve the mapping. ARP is transmitted to any core switch, which in turn distributes it to all pods and finally all edge switches.  Less switch states requirement  VM migration Forwarding an invalidation message. Edge switch traps the packets.

 LDP: Location Discovery Protocol  LDM: Location Discovery Message switch identifier pod pos Level: 0, 1, 2 up/down

 Switch updates the set of switch neighbors that it has heard from.  If a switch is not connected to more than k/2 neighbor switches for sufficiently long, it concludes that it is an edge switch  On any subsequent LDM it receives, it infers that the corresponding incoming port is an upward facing one.  Conform its notion of position

 Point out aggregation switch and corresponding downward facing port  Set the direction of some of edge/aggregation switches ports  Determining the level for core switches

 Edge switches must acquire a unique position number in each pod  Pod number spreads to the rest of the pod

 Once switches establish their local positions using LDP, they employ updates from their neighbors to populate their forwarding tables.  Packets will always be forwarded up to either an aggregation or core switch and then down toward their ultimate destination.  Once a packet begins to travel down, it is not possible for it to travel back up the topology.

 Convergence time with increasing faults

 TCP convergence

 Multicast convergence

 Scalability

 VM Migration

 Fat Tree is a suitable topology for datacenter.  PMAC and new protocols make fabric work efficiently.  Portland makes datacenter networks more flexible, efficient, fault tolerant and easy to manage.  VM migration is easier in the new fabric

 Overhead?  Load of fabric manager?  Easy to implement?