• Home
• Syllabus
• Schedule
• Connect
• Checkin
• Homework
  • HW0
  • HW1
  • HW2
  • HW3
  • HW4
  • HW5
  • HW6
  • HW7
• Piazza
• Slides
• Other Info

See this page as a slide show

CT320 Routing

Thanks to:

• Dr. Indrajit Ray, CSU
• Dr. James Walden, NKU
• Russ Wakefield, CSU

for the contents of these slides.

The Internet Network layer

Remember the layers?

• Application Layer
• Transport Layer
• Network Layer
• Data Link Layer

The Internet Network layer

Host, router network layer functions:

• IP protocol
  • addressing conventions
  • datagram format
  • packet handling conventions
• Routing protocols
  • Path selection
  • RIP, OSPF, BGP
• ICMP protocol
  • error reporting
  • router “signaling”

IP Fragmentation & Reassembly

• Like a big text message
• Network links have MTU (max. transmission unit) largest possible link-level frame.
  • different link types, different MTUs
• Large IP datagram divided (“fragmented”) within net
  • one datagram becomes several datagrams
  • “reassembled” only at final destination
  • IP header bits used to identify, order related fragments
• Fragmentation:
  • in: one large datagram
  • out: several smaller datagrams

IP Fragmentation and Reassembly Example

• 4000 byte datagram
• MTU = 1500 bytes (1480 bytes of data)

Original:

Fragments:

(185 = 1480 / 8)

Joke

(Well, it’s the setup to a joke.)

Hierarchical Addressing: Route Aggregation

How it’s not done:

┌─────────────┐       ┌────────┐    ┌────────┐       ┌──────────────┐
│ 192.0.2.4   │·······│ ISP #1 │    │ ISP #2 │·······│ 203.0.113.42 │
└─────────────┘   :   └────────┘    └────────┘   :   └──────────────┘
┌─────────────┐   :      :                :      :   ┌──────────────┐
│ 203.0.113.8 │···:      :  ┌──────────┐  :      :···│ 192.0.2.88   │
└─────────────┘   :      ···│ Internet │···      :   └──────────────┘
┌─────────────┐   :         └──────────┘         :   ┌──────────────┐
│ 192.0.2.66  │····               :              ····│ 203.0.113.17 │
└─────────────┘                   :                  └──────────────┘

• ISP #1: “Internet, forward 192.0.2.4, 203.0.113.8, and 192.0.2.66 to me.”
• ISP #2: “Internet, forward 203.0.113.42, 192.0.2.88, and 203.0.113.17 to me.”
• Imagine if each ISP had thousands of customers!

Hierarchical Addressing: Route Aggregation

How it’s done:

┌────────────┐       ┌────────┐    ┌────────┐       ┌──────────────┐
│ 192.0.2.4  │·······│ ISP #1 │    │ ISP #2 │·······│ 203.0.113.42 │
└────────────┘   :   └────────┘    └────────┘   :   └──────────────┘
┌────────────┐   :      :                :      :   ┌──────────────┐
│ 192.0.2.88 │···:      :  ┌──────────┐  :      :···│ 203.0.113.8  │
└────────────┘   :      ···│ Internet │···      :   └──────────────┘
┌────────────┐   :         └──────────┘         :   ┌──────────────┐
│ 192.0.2.66 │····               :              ····│ 203.0.113.17 │
└────────────┘                   :                  └──────────────┘

• ISP #1: “Internet, forward 192.0.2/24 to me.”
• ISP #2: “Internet, forward 203.0.113/24 to me.”

This scales nicely.

Hierarchical Addressing: More Specific Routes

┌────────────┐     ┌────────┐  ┌────────┐······················
│ 192.0.2.4  │·····│ ISP #1 │  │ ISP #2 │                     :
└────────────┘  :  └────────┘  └────────┘·:  ┌──────────────┐ :  ┌────────────┐
┌────────────┐  :    :               :    :··│ 203.0.113.42 │ ···│ 192.0.2.11 │
│ 192.0.2.88 │··:    :  ┌──────────┐ :    :  └──────────────┘ :  └────────────┘
└────────────┘  :    ···│ Internet │··    :  ┌──────────────┐ :  ┌────────────┐
┌────────────┐  :       └──────────┘      :··│ 203.0.113.8  │ :··│ 192.0.2.22 │
│ 192.0.2.66 │···             :           :  └──────────────┘ :  └────────────┘
└────────────┘                :           :  ┌──────────────┐ :  ┌────────────┐
                                          ···│ 203.0.113.17 │ ···│ 192.0.2.97 │
                                             └──────────────┘    └────────────┘

• ISP #1: “Internet, forward 192.0.2/24 to me.”
• ISP #2: “Internet, forward 203.0.113/24 and 192.0.2/24 to me.”

Hierarchical Addressing

• Without hierarchical addressing, top-level routers would have to know the location of every host in the world.
• With hierarchical addressing, top-level routers only have to know where big hunks of hosts are.
• It’s a lot easier to treat all of CSU as one entity, as opposed to having sixty-five thousand entries for all the possible CSU hosts.
• Now, imagine how many hosts NASA has. Or HP. Or the Pentagon.
• Phone call routing used to work this way, until everything got all mobile.

IP Addressing: ICANN

• How does an ISP get block of addresses?
• ICANN: Internet Corporation for Assigned Names and Numbers
  • allocates addresses
  • manages DNS
  • assigns domain names, resolves disputes

Addressing at CSU

• All of CSU is 129.82/16
• That’s more than sixty-five thousand possible hosts.
• Does the CSU main router have to have sixty-five thousand entries?
• This is the same problem as we have at the top level of the internet.
• So, use the same solution!

Hierarchy at CSU

• Hierarchy saves the day again!
• Groups within CSU get blocks within CSU’s 129.82/16:
  • 129.82.28/22 Morgan Library
  • 129.82.40/22 College of Business
  • 129.82.44/23 Computer Science
  • 129.82.48/22 Atmospheric Science
  • 129.82.76/22 Chemistry
  • 129.82.220/22 Lory Student Center
  • 129.82.224/21 Engineering (more than CS!?)
    • wait—how is /21 more than /23?
  • and scores more
• ICANN neither knows nor cares about this.

Hierarchy at CSU

┌───────────────┐     ┌──────────┐        ┌────────┐     ┌───────────────┐
│ 129.82.40.123 │·····│ Business │        │   CS   │·····│ 129.82.44.12  │
└───────────────┘  :  └──────────┘        └────────┘  :  └───────────────┘
┌───────────────┐  :      :   ┌────────────┐   :      :  ┌───────────────┐
│ 129.82.43.4   │··:      ····│ CSU router │····      :··│ 129.82.45.85  │
└───────────────┘  :          └────────────┘          :  └───────────────┘
┌───────────────┐  :                :                 :  ┌───────────────┐
│ 129.82.42.66  │···          ┌────────────┐          ···│ 129.82.45.234 │
└───────────────┘             │    FRGP    │             └───────────────┘
                              └────────────┘ 

• The CSU router doesn’t need 65 000-some entries, just one per subnet: one for Business, one for CS, one for Engineering, etc.
• When the Business School adds another computer, the CSU router doesn’t have to be updated, since the new computer is within 129.82.40/22.
• CSUs connects to the Internet via Front Range GigaPop, costing >$160 000 in 2017.

IP datagram format, redux

ICMP: Internet Control Message Protocol

• used by hosts & routers to communicate network-level information
  • error reporting: unreachable host, network, port, protocol
  • echo request/reply (used by ping)
• network-layer “above” IP:
  • ICMP msgs carried in IP datagrams
• ICMP message: type, code plus first 8 bytes of IP datagram causing error

Ping

• Create an IP packet with type=8, code=0.
• Send it to the destination
• Destination will create an IP packet with type=0, code=0, and send it back.
  • If it feels like it.

Ping examples

$ ping -c1 google.com
PING google.com (142.250.69.238) 56(84) bytes of data.
64 bytes from den08s05-in-f14.1e100.net (142.250.69.238): icmp_seq=1 ttl=57 time=3.13 ms

--- google.com ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 3.129/3.129/3.129/0.000 ms

$ ping -c1 tuba
PING tuba.cs.colostate.edu (10.1.44.62) 56(84) bytes of data.
64 bytes from tuba.cs.colostate.edu (10.1.44.62): icmp_seq=1 ttl=63 time=2.99 ms

--- tuba.cs.colostate.edu ping statistics ---
1 packets transmitted, 1 received, 0% packet loss, time 0ms
rtt min/avg/max/mdev = 2.990/2.990/2.990/0.000 ms

Traceroute and ICMP

• Source sends series of UDP packets to dest
  • First has TTL=1, second has TTL=2, third has TTL=3, etc.
  • Unlikely port number
• When nth datagram arrives at the nth router:
  • Router discards datagram, because we ran out of TTL hops
  • Sends back an ICMP message packet (type 11, code 0), “too many hops”
  • Message includes name of source router & IP address
• When a packet actually gets to the destination:
  • Destination returns ICMP “port unreachable” (type 3, code 3)
  • When source gets this ICMP, stops.
• This all assumes that routing is consistent—what about anycast?

User: Guest