031_DN22_VL16_RouterArchitecture

Router Architecture Overview
What’s inside a router?

Prof. Anja Feldmann, Ph.D.
What does a Router Look Like?
Ericsson SSR 8020 Cisco CRS-1 Juniper T4000

BNG/BRAS/PGW Core Router Core Router
- Maximum 16 Tbit/s - 2.2Tbit/s single chassis - 3.8 Tbit/s single chassis
2
- Up to 322Tbit/s for multichassis
And like this:
Dlink DIR-615 Wireless N 300 Belkin (formerly Cisco, Linksys)

Home Router N600 DB Wireless Dual-Band N+
- LAN: 4x 10/100Mbit/s Ports Home Router
- WAN: 1x 10/100Mbit/s Port - LAN: 4x 10/100Mbit/s Ports
- Up to 300 Mbit/s throughput - WAN: 1x 10/100Mbit/s Port
- WiFi support - Up to 300 Mbit/s throughput
- WiFi support
Data Networks Router Architecture 3
Who Makes Core Routers?
• Cisco
• CRS (Carrier Router Series)
• Juniper
• T-series
• Nokia (used to be Alcatel-Lucent)
• XRS (Extensible Routing System)
• Huawei
• Netengine
• Others manufacture aggregation/access
networking gear, e.g., for edge deployments
Router architecture overview
Two key router functions:
• Run routing algorithms/protocol (RIP, OSPF, BGP)
• Forwarding datagrams from incoming to outgoing link Incoming reachability information via
RIP/OSPF/IS-IS/BGP used to compute
routing information Base (RIB)
Forwarding information base routing
(FIB) computed, routing, management
pushed to line cards processor
control plane (software)
forwarding data
plane (hardware)
high-speed
switching
fabric
router line cards on router line cards on

input side Router Architecture output side 5
RIB (Routing Information Base)
• Router can contain many different RIBs
• One for each routing protocol
• Usually consolidated into one global RIB or into FIB
• (End system IP addresses (/32s) populated through ARP for
default gateway MAC)
• Minimum contents
• Network id of destination subnet
• Cost/metric for hop
• Next hop gateway or end system
• Other information
• Quality of service, e.g., a U if the link is up
• Access control lists for security
• Interface, such as eth0 for first Ethernet line card, etc.
RIB
• Router can contain many different RIBs
• Minimum contents
• Other information
Loopback
metric is
IP address of low
Network + Netmask= nework id Next hop
(192.168.0.0/24 in this case next hop interface
7
Input port functions
lookup,
link forwarding
line layer switch
termination protocol fabric
(receive)
queueing
Physical layer:
Bit-level reception
Data link layer: Decentralized switching:
E.g., Ethernet • Given datagram dest., lookup output port using
see chapter 5 forwarding table in input port memory (“match plus
action”)
• Goal: Complete input port processing at ‘line speed’
• Queuing: If datagrams arrive faster than forwarding
Data Networks rate into switch fabric 8
Switching fabrics
• Transfer packet from input buffer to appropriate output buffer
• Switching rate: Rate at which packets can be transfer from inputs
to outputs
• Often measured as multiple of input/output line rate
• N inputs: Switching rate N times line rate desirable
• Three types of switching fabrics
memory
memory bus dedicated

fabric
Switching via memory
First generation routers:
• Traditional computers with switching under direct control of CPU
• Packet copied to system’s memory
• Speed limited by memory bandwidth (2 bus crossings per datagram)
input output
port memory port
(e.g., (e.g.,
Ethernet) Ethernet)
system bus

Switching via a bus
Datagram is switched
➡ From input port memory
⬅ To output port memory
via a shared bus
bus
Bus contention: Switching speed limited by
bus bandwidth
E.g. Cisco 5600: 32 Gbps bus
Sufficient speed for access and enterprise routers

Switching via Dedicated Fabric
• Overcome bus bandwidth limitations
• Banyan networks, crossbar, other interconnection nets initially
developed to connect
processors in multiprocessor
• Advanced design:
• Fragmenting datagram into
fixed length cells crossbar
• Append hardware address of
output line card to front of cell
• Switch cells through the fabric.
• Cisco 12000: Switches 60 Gbps via the interconnection network

Output ports
datagram
switch buffer link
fabric layer line
protocol termination
(send)
queueing
Buffering: Required when datagrams arrive from

fabric faster than the transmission rate
Scheduling discipline: Chooses among queued
datagrams for transmission
Output port queueing
switch
switch
fabric
fabric
at t, packets move one packet per time

from input to output
Buffering when arrival rate via switch exceeds output line speed
Queueing (delay) and loss due to output port buffer overflow!

How much buffering?
• RFC 3439 rule of thumb: Average buffering:
“typical” RTT times link capacity C
• For C = 10 Gpbs link, RTT = 250 msec:
2.5 Gbit buffer
• Recent recommendation:
With N flows, buffering equal to: RTT. C
N

Input port queuing
• Fabric slower than input ports combined -> queueing at input queues
• Queueing delay and loss due to input buffer overflow!
• Head-of-the-Line (HOL) blocking: Queued datagram at front of queue
prevents others in queue from moving forward
switch switch
fabric fabric
output port contention: one packet time

Only one red datagram can be later: Green packet
transferred. experiences HOL
lower red packet is blocked blocking
Summary
• Routers have evolved from dedicated workstations/PCs
to heavy duty industrial equipment costing millions of €s
• Capable of switching 300+ Tb/s
• Dedicated line cards separate fast path from slow path
• Slow path through route processor primarily for control
plane traffic
• Different hardware approaches to accelerating fast path
forwarding
• Queuing and buffering can help or hinder traffic flow

031_DN22_VL16_RouterArchitecture

Uploaded by

Copyright:

Available Formats

031_DN22_VL16_RouterArchitecture

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

031_DN22_VL16_RouterArchitecture

Uploaded by

Copyright:

Available Formats

Router Architecture Overview

What’s inside a router?

Ericsson SSR 8020 Cisco CRS-1 Juniper T4000

Dlink DIR-615 Wireless N 300 Belkin (formerly Cisco, Linksys)

router line cards on router line cards on

memory bus dedicated

Data Networks Router Architecture 10

Data Networks Router Architecture 11

Data Networks Router Architecture 12

Buffering: Required when datagrams arrive from

at t, packets move one packet per time

Data Networks Router Architecture 14

Data Networks Router Architecture 15

output port contention: one packet time

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.