Transcript Chapter 1 - It works!

Comp 365
Computer Networks
Fall 2014
Chapter 1
Part 2
Network Core
These slides derived from Computer Networking: A
Top Down Approach ,
6th edition.
Jim Kurose, Keith Ross
Addison-Wesley, March 2012.
Introduction
1-1
Chapter 1: roadmap
1.1 What is the Internet?
1.2 Network edge
 end systems, access networks, links
1.3 Network core
 circuit switching, packet switching, network structure
1.4 Delay, loss and throughput in packet-switched
networks
1.5 Protocol layers, service models
1.6 Networks under attack: security
1.7 History
Introduction
1-2
The Network Core
 mesh of interconnected
routers
 the fundamental
question: how is data
transferred through net?
 circuit switching:
dedicated circuit per
Resource
s
call: telephone net
reserved
 packet-switching: data
sent thru net in
Resources
allocated
discrete “chunks”
on demand.
Core Resources: buffers, link transmission rate
Introduction
1-3
The Network Core
 The internet is packet
switched
 Telephone network is
circuit switched
 We need to understand
circuit switching to know
why it’s not used for the
internet
Introduction
1-4
Network Core: Circuit Switching
End-end resources
reserved for “call”
 link bandwidth, switch





capacity
dedicated resources: no
sharing
circuit-like (guaranteed)
performance
call setup required
State maintained
Data transferred at a
guaranteed rate
Introduction
1-5
Network Core: Circuit Switching
End-end resources
reserved for “call”
 Links between circuit
switches
 Each link can support n
circuits
 There can be n
simultaneous connections.
 Each circuit thus gets 1/n
of the link’s bandwidth.
Introduction
1-6
Network Core: Circuit Switching
network resources
(e.g., bandwidth)
divided into “pieces”
 pieces allocated to calls
 dividing link bandwidth
into “pieces”
 frequency division
 time division
 resource piece idle if
not used by owning call
(no sharing)
 When two hosts want to
communicate network
establishes a dedicated
end-to-end connection
between the hosts.
Introduction
1-7
Alternative core: circuit switching
end-end resources allocated to,
reserved for “call” between
source & dest:
In diagram, each link has four circuits.
 call gets 2nd circuit in top link and
1st circuit in right link.
 dedicated resources: no sharing
 circuit-like (guaranteed)
performance
 circuit segment idle if not used by call
(no sharing)
 Commonly used in traditional
telephone networks

Introduction
1-8
Circuit Switching: FDM and TDM
Example:
FDM
4 users
Each connection
gets one band
of frequencies
frequency
FDM: frequency-division multiplexing
time
TDM
Each connection
gets one time
slot
frequency
TDM: time-division multiplexing
time
Introduction
1-9
Circuit Switching: FDM and TDM
Example: telephone networks
FDM
4 users
4kHz
4kHz
4kHz
4kHz
frequency
time
FDM: frequency-division multiplexing
4kHz = 4,000 hertz = 4,000 cycles per second
Radio stations also use FDM to share the frequency spectrum between 88 MHz and 108 MHz
Introduction
1-10
Circuit Switching: FDM and TDM
Example:
4 users
TDM: time-division multiplexing
Circuit gets same time slot in every frame
TDM
One time slot
frequency
One frame
time
Transmission rate of a circuit: frame rate multiplied by the number of bits in a slot.
Example: 8,000 frames per sec, slot has 8 bits, then what is the transmission rate?
64 kbps
Introduction
1-11
Numerical example
 How long does it take to send a file of
640,000 bits from host A to host B over a
circuit-switched network?
All links have transmission rate of 1.536 Mbps
 Each link uses TDM with 24 slots/sec, 24
slots/frame.
 500 msec to establish end-to-end circuit

Let’s work it out!
Introduction
1-12
Numerical example
 How long does it take to send a file of 640,000 bits from
host A to host B over a circuit-switched network?



All links have transmission rate of 1.536 Mbps
Each link uses TDM with 24 slots/sec, 24 slots/frame.
500 msec to establish end-to-end circuit
 Soln:
 determine how many slots we need. To do this must figure out
how many bits/slot:
(1.536 x 106 bits/sec) / (24 slots/sec) = 64,000 bits / slot

Now can determine how many slots we need:
640,000 bits / (64,000 bits/slot) = 10 slots.


So need 10 seconds + 500 msec = 10.5 seconds.
This does not account for propagation delays.
Introduction
1-13
Circuit switching: analysis
 Disadvantages:
Network resources are wasted during “silent”
times (when no one is talking but still connected)
 Establishing end-to-end circuits and reserving
end-to-end bandwidth is complicated and requires
complex signaling software.

 Advantage:
 Guaranteed bandwith and transmission time
 Necessary for some applications (streamed
music/video)
Introduction
1-14
Network Core: Packet Switching
transmission:
 Each end-end data stream divided into packets
 Each packet travels through communication links
 Links connected by packet switches
 Routers
 Or link-level switches.
 Switches use store-and-forward transmission
 Switch receives entire packet before it transmits any of it again
Introduction
1-15
Network Core: Packet Switching
transmission:
Introduction
1-16
Network Core: Packet Switching
each end-end data stream
divided into packets
 user A, B packets share
network resources
 each packet uses full link
bandwidth
 resources used as needed
Bandwidth division into “pieces”
Dedicated allocation
Resource reservation
Introduction
1-17
Network Core: Packet Switching
Switching delays
 Assume packet has L bits
 Assume there are Q switches
 Assume a switch transmits at rate of R
 Takes each switch how long to transmit the packet?

L/R
 Total delay is:
 QL/R
Introduction
1-18
Packet-switching: store-and-forward
L
R
R
This Figure:
 takes L/R seconds to
transmit (push out)
packet of L bits on to
link at R bps
 store and forward:
entire packet must
arrive at router before
it can be transmitted
on next link
 delay = 3L/R (assuming
zero propagation delay)
R
Example:
 L = 7.5 Mbits
 R = 1.5 Mbps
 transmission delay = ??

15 sec
more on delay shortly …
Introduction
1-19
Network Core: Packet Switching
Switching delays:
 Each switch has multiple links
 Each link has buffer
 If packet arrives and another packet is already being
transmitted on that link, must wait in queue
 Called queuing delay.
 Varies depending on network congestion
 Packet loss: queue is full when packet arrives.
Introduction
1-20
Packet Switching: Statistical Multiplexing
100 Mb/s
Ethernet
A
B
statistical multiplexing
C
1.5 Mb/s
queue of packets
waiting for output
link
D
On-demand sharing of resources is
called statistical multiplexing
E
Sequence of A & B packets does not have fixed pattern,
bandwidth shared on demand  statistical multiplexing.
TDM: each host gets same slot in revolving TDM frame.
Introduction
1-21
Two key network-core functions
routing: determines sourcedestination route taken by
packets
 routing algorithms
forwarding: move packets from
router’s input to appropriate
router output
routing algorithm
local forwarding table
header value output link
0100
0101
0111
1001
3
2
2
1
1
3 2
dest address in arriving
packet’s header
Network Layer
4-22
Packet switching versus circuit switching
Packet switching allows more users to use network!
 1 Mb/s link
 For each user assume:
 100 kb/s when “active”
 active 10% of time
 circuit-switching:
 10 users
 packet switching:
 with 35 users,
probability > 10 active
at same time is less
than .0004
N users
1 Mbps link
Q: how did we get value 0.0004?
A: see this pdf:
https://147.129.181.14/~barr/classes/comp3
65/lectures/Prob10orMoreUsers.pdf
Introduction
1-23
Packet switching versus circuit switching
Packet switching allows more users to use network!
 packet switching:
 with 35 users, probability > 10 active at same time is less
than .0004
 With 10 or fewer simultaneously active users, aggregate
arrival rate of data is less than or equal to 1 Mbps.
 No loss of time
 Allows essentially same performance as circuit switching
 But allows > 3 times number of users
Introduction
1-24
Packet switching versus circuit switching
Is packet switching a “slam dunk winner?”
 great for bursty data
Most networks use packet-switching; even
telephone networks are going to this!
resource sharing
 simpler, no call setup
 excessive congestion: packet delay and loss
 protocols needed for reliable data transfer,
congestion control
 Q: How to provide circuit-like behavior?
 bandwidth guarantees needed for audio/video apps
 still an unsolved problem (chapter 7)

Q: human analogies of reserved resources (circuit
switching) versus on-demand allocation (packet-switching)?
Introduction
1-25
Routing
 How do packets make their way through
packet-switched Networks?
Introduction
1-26
Internet structure: network of networks




End systems connect to Internet via access ISPs (Internet
Service Providers)
 Residential, company and university ISPs
Access ISPs in turn must be interconnected.
 So that any two hosts can send packets to each other
Resulting network of networks is very complex
 Evolution was driven by economics and national policies
Let’s take a stepwise approach to describe current Internet
structure
Internet structure: network of networks
Question: given millions of access ISPs, how to connect them
together?
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
Internet structure: network of networks
Option: connect each access ISP to every other access ISP?
access
net
access
net
access
net
access
net
access
net
access
net
access
net
connecting each access ISP
to each other directly doesn’t
scale: O(N2) connections.
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
Internet structure: network of networks
Option: connect each access ISP to a global transit ISP? Customer
and provider ISPs have economic agreement.
access
net
access
net
access
net
access
net
access
net
access
net
access
net
global
ISP
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
access
net
Internet structure: network of networks
But if one global ISP is viable business, there will be competitors
….
access
net
access
net
access
net
access
net
access
net
access
net
access
net
ISP A
access
net
access
net
access
net
ISP B
ISP C
access
net
access
net
access
net
access
net
access
net
access
net
Internet structure: network of networks
But if one global ISP is viable business, there will be competitors
…. which must be interconnected
Internet exchange point
access
access
net
net
access
net
access
net
access
net
IXP
access
net
ISP A
IXP
access
net
access
net
access
net
access
net
ISP B
ISP C
access
net
peering link
access
net
access
net
access
net
access
net
access
net
Internet structure: network of networks
… and regional networks may arise to connect access nets to
ISPS
access
net
access
net
access
net
access
net
access
net
IXP
access
net
ISP A
IXP
access
net
access
net
access
net
access
net
ISP B
ISP C
access
net
access
net
regional net
access
net
access
net
access
net
access
net
Internet structure: network of networks
… and content provider networks (e.g., Google, Microsoft,
Akamai ) may run their own network, to bring services, content
close to end users
access
net
access
net
access
net
access
net
access
net
IXP
access
net
ISP A
access
net
Content provider network
IXP
access
net
access
net
access
net
ISP B
ISP B
access
net
access
net
regional net
access
net
access
net
access
net
access
net
Internet structure: network of networks
Tier 1 ISP
Tier 1 ISP
IXP
IXP
Regional ISP
access
ISP

access
ISP
Google
access
ISP
access
ISP
IXP
Regional ISP
access
ISP
access
ISP
access
ISP
access
ISP
at center: small # of well-connected large networks


“tier-1” commercial ISPs (e.g., Level 3, Sprint, AT&T, NTT), national &
international coverage
content provider network (e.g, Google): private network that connects it data
Introduction
centers to Internet, often bypassing tier-1, regional ISPs
1-35
Tier-1 ISP: e.g., Sprint
POP: point-of-presence
to/from backbone
peering
…
…
…
…
…
to/from customers
Introduction
1-36