0% found this document useful (0 votes)
2 views

Lecture 2_CSC523_Computer Networks

The document provides an overview of the Internet's structure, including concepts like packet and circuit switching, routing, and performance metrics such as loss, delay, and throughput. It explains the network core functions, the differences between packet-switching and circuit-switching, and the complexities of connecting numerous Internet Service Providers (ISPs). Additionally, it discusses packet delay sources, queuing, and loss, emphasizing the importance of managing traffic intensity to ensure efficient network performance.

Uploaded by

teeboibass
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2 views

Lecture 2_CSC523_Computer Networks

The document provides an overview of the Internet's structure, including concepts like packet and circuit switching, routing, and performance metrics such as loss, delay, and throughput. It explains the network core functions, the differences between packet-switching and circuit-switching, and the complexities of connecting numerous Internet Service Providers (ISPs). Additionally, it discusses packet delay sources, queuing, and loss, emphasizing the importance of managing traffic intensity to ensure efficient network performance.

Uploaded by

teeboibass
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 37

Lecture 1: roadmap

 What is the Internet?


 What is a protocol?
 Network edge: hosts, access
network, physical media
 Network core: packet/circuit
switching, internet structure
 Performance: loss, delay, throughput
 Security
 Protocol layers, service models
 History
Introduction: 1-1
The network core
 mesh of interconnected routers mobile network
national or global ISP
 packet-switching: hosts break
application-layer messages into packets
• network forwards packets from one
router to the next, across links on local or
path from source to destination regional ISP

home network content


provider
network datacenter
network

enterprise
network

Introduction: 1-2
Two key network-core functions

routing algorithm Routing:


 global action:
Forwarding: local forwarding table
header value output link determine source-
 aka “switching” 0100
0101
3
2 destination paths
 local action: 0111
1001
2
1 taken by packets
move arriving  routing algorithms
packets from
1
router’s input link
to appropriate 3 2
router output link 011
1

destination address in arriving


packet’s header
Introduction: 1-3
Packet-switching: store-and-
forward
L bits
per packet
3 2 1
source destination
R bps R bps

 packet transmission delay: takes L/R seconds to One-hop numerical example:


transmit (push out) L-bit packet into link at R bps  L = 10 Kbits
 store and forward: entire packet must arrive at  R = 100 Mbps
router before it can be transmitted on next link  one-hop transmission delay
= 0.1 msec

Introduction: 1-4
Packet-switching: queueing
R = 100 Mb/s
A C

D
B R = 1.5 Mb/s
E
queue of packets
waiting for transmission
over output link

Queueing occurs when work arrives faster than it can be serviced:

Introduction: 1-5
Packet-switching: queueing
R = 100 Mb/s
A C

D
B R = 1.5 Mb/s
E
queue of packets
waiting for transmission
over output link

Packet queuing and loss: if arrival rate (in bps) to link exceeds
transmission rate (bps) of link for some period of time:
 packets will queue, waiting to be transmitted on output link
 packets can be dropped (lost) if memory (buffer) in router fills up
Introduction: 1-6
Alternative to packet switching: circuit
switching
end-end resources allocated to,
reserved for “call” between source
and destination
 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

* Check out the online interactive exercises for more examples: h ttp://gaia.cs.umass.edu/kurose_ross/interactive
Packet switching versus circuit
switching
example:
 1 Gb/s link

…..
N
 each user: users 1 Gbps link
• 100 Mb/s when “active”
• active 10% of time
Q: how many users can use this network under circuit-switching and packet switching?

 circuit-switching: 10 users
 packet switching: with 35 users, Q: how did we get value 0.0004?
probability > 10 active at same time A: HW problem (for those with
is less than .0004 * course in probability only)

* Check out the online interactive exercises for more examples: h ttp://gaia.cs.umass.edu/kurose_ross/interactive
Introduction: 1-8
Packet switching versus circuit
switching
Is packet switching a “slam dunk winner”?
 great for “bursty” data – sometimes has data to send, but at other times not
• resource sharing
• simpler, no call setup
 excessive congestion possible: packet delay and loss due to buffer overflow
• protocols needed for reliable data transfer, congestion control
 Q: How to provide circuit-like behavior with packet-switching?
• “It’s complicated.” We’ll study various techniques that try to make packet
switching as “circuit-like” as possible.

Introduction: 1-9
Internet structure: a “network of networks”
mobile network
 hosts connect to Internet via access national or global ISP
Internet Service Providers (ISPs)
 access ISPs in turn must be
interconnected
• so that any two hosts (anywhere!) local or
regional ISP
can send packets to each other
 resulting network of networks is home network content
provider
very complex network datacenter
network

• evolution driven by economics, enterprise


national policies network

Let’s take a stepwise approach to describe current Internet structure


Internet structure: a “network of networks”
Question: given millions of access ISPs, how to connect them together?

access
… access
net
access
net …
net
access
access net
net
access
access net
net


access access
net net

access
net
access
net

access
net
access
… net
access access …
net access net
net

Introduction: 1-11
Internet structure: a “network of networks”
Question: given millions of access ISPs, how to connect them together?

access
… access
net
access
net …
net
access
access
net
… … net

access
access net
net

connecting each access ISP to



each other directly doesn’t scale:


O(N2) connections.
access access

net net

access
net
access
net

access
net
access
… access
… … net
access
net access net
net

Introduction: 1-12
Internet structure: a “network of networks”
Option: connect each access ISP to one global transit ISP?
Customer and provider ISPs have economic agreement.

access
… access
net
access
net …
net
access
access net
net
access
access net
net


global
access
net
ISP access
net

access
net
access
net

access
net
access
… net
access access …
net access net
net

Introduction: 1-13
Internet structure: a “network of networks”
But if one global ISP is viable business, there will be competitors ….

access
… access
net
access
net …
net
access
access net
net
access
access net
net
ISP A


access
net ISP B access
net

access
net
ISP C
access
net

access
net
access
… net
access access …
net access net
net

Introduction: 1-14
Internet structure: a “network of networks”
But if one global ISP is viable business, there will be competitors …. who will
want to be connected
Internet exchange point
access
… access
net
access
net …
net
access
access net
net
IXP access
access net
net
ISP A


access
net
IXP ISP B access
net

access
net
ISP C
access
net

access
net
peering link
access
… net
access access …
net access net
net

Introduction: 1-15
Internet structure: a “network of networks”
… and regional networks may arise to connect access nets to ISPs

access
… access
net
access
net …
net
access
access net
net
IXP access
access net
net
ISP A


access
net
IXP ISP B access
net

access
net
ISP C
access
net

access
net
regional ISP access
… net
access access …
net access net
net

Introduction: 1-16
Internet structure: a “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
… access
net
access
net

net
access
access net
net
IXP access
access net
net
ISP A


Content provider network
access
net
IXP ISP B access
net

access
net
ISP C
access
net

access
net
regional ISP access
… net
access access …
net access net
net

Introduction: 1-17
Internet structure: a “network of networks”

Tier 1 ISP Tier 1 ISP Google


IXP IXP IXP
Regional ISP Regional ISP

access access access access access access access access


ISP ISP ISP ISP ISP ISP ISP 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 networks (e.g., Google, Facebook): private network that connects its
data centers to Internet, often bypassing tier-1, regional ISPs
Introduction: 1-18
Lecture 1: roadmap
 What is the Internet?
 What is a protocol?
 Network edge: hosts, access
network, physical media
 Network core: packet/circuit
switching, internet structure
 Performance: loss, delay, throughput
 Security
 Protocol layers, service models
 History
Introduction: 1-19
How do packet delay and loss occur?
 packets queue in router buffers, waiting for turn for transmission
 queue length grows when arrival rate to link (temporarily) exceeds output link
capacity
 packet loss occurs when memory to hold queued packets fills up
packet being transmitted (transmission delay)

B
packets in buffers (queueing delay)
free (available) buffers: arriving packets
dropped (loss) if no free buffers
Introduction: 1-20
Packet delay: four sources
transmission
A propagation

B
nodal
processing queueing

dnodal = dproc + dqueue + dtrans + dprop

dproc: nodal processing dqueue: queueing delay


 check bit errors  time waiting at output link for
 determine output link transmission
 typically < microsecs  depends on congestion level of
router
Introduction: 1-21
Packet delay: four sources
transmission
A propagation

B
nodal
processing queueing

dnodal = dproc + dqueue + dtrans + dprop


dtrans: transmission delay: dprop: propagation delay:
 L: packet length (bits)  d: length of physical link
 R: link transmission rate (bps)  s: propagation speed (~2x108 m/sec)
 dtrans = L/R  dprop = d/s
dtrans and dprop
very different Introduction: 1-22
Caravan analogy
100 km 100 km

ten-car caravan toll booth toll booth toll booth


(aka 10-bit packet) (aka link)

 car ~ bit; caravan ~ packet; toll  time to “push” entire caravan


service ~ link transmission through toll booth onto
 toll booth takes 12 sec to service highway = 12*10 = 120 sec
car (bit transmission time)  time for last car to propagate
 “propagate” at 100 km/hr from 1st to 2nd toll both:
100km/(100km/hr) = 1 hr
 Q: How long until caravan is lined
up before 2nd toll booth?  A: 62 minutes

Introduction: 1-23
Caravan analogy
100 km 100 km

ten-car caravan toll booth toll booth


(aka 10-bit packet) (aka router)

 suppose cars now “propagate” at 1000 km/hr


 and suppose toll booth now takes one min to service a car
 Q: Will cars arrive to 2nd booth before all cars serviced at first booth?
A: Yes! after 7 min, first car arrives at second booth; three cars still at
first booth

Introduction: 1-24
Packet queueing delay (revisited)
 a: average packet arrival rate

average queueing delay


 L: packet length (bits)
 R: link bandwidth (bit transmission rate)

L .a arrival rate of bits “traffic


:
R service rate of bits intensity” traffic intensity = La/R 1

 La/R ~ 0: avg. queueing delay small La/R ~ 0

 La/R -> 1: avg. queueing delay large


 La/R > 1: more “work” arriving is
more than can be serviced - average
delay infinite!
Golden rules in traffic engineering is: Design your system so that the traffic intensity is no La/R -> 1
greater than 1.
Packet loss
 queue (aka buffer) preceding link in buffer has finite capacity
 packet arriving to full queue dropped (aka lost)
 lost packet may be retransmitted by previous node, by source end
system, or not at all
buffer
(waiting area) packet being transmitted
A

B
packet arriving to
full buffer is lost

* Check out the Java applet for an interactive animation (on publisher’s website) of queuing and loss
Introduction: 1-26
Throughput
 throughput: rate (bits/time unit) at which bits are being sent from
sender to receiver
• instantaneous: rate at given point in time
• average: rate over longer period of time

link
pipecapacity
that can carry linkthat
pipe capacity
can carry
serverserver,
sends with
bits Rsfluid at rate
bits/sec Rfluid
c
at rate
bits/sec
(fluid)
fileinto
of Fpipe
bits (Rs bits/sec) (Rc bits/sec)
to send to client
Introduction: 1-27
Throughput
Rs < Rc What is average end-end throughput?

Rs bits/sec Rc bits/sec

Rs > Rc What is average end-end throughput?

Rs bits/sec Rc bits/sec

bottleneck link
link on end-end path that constrains end-end throughput
Introduction: 1-28
Throughput
Throughput
Throughput: network scenario
 per-connection end-end
Rs throughput:
Rs Rs min(Rc,Rs,R/10)
 in practice: Rc or Rs is
R
often bottleneck
Rc Rc
Rc
* Check out the online interactive exercises for more
examples:
http://gaia.cs.umass.edu/kurose_ross/interactive/
10 connections (fairly) share
backbone bottleneck link R bits/sec
Introduction: 1-31
Services, Layering and Encapsulation
M

application application
Ht M
transport Transport-layer protocol transfers M (e.g., reliably) from transport
one process to another, using services of network layer
Hn Ht M
network Network-layer protocol transfers transport-layer segment
network
[Ht | M] from one host to another, using link layer services
link  network-layer protocol encapsulates link
transport-layer segment [Ht | M] with
physical network layer-layer header Hn to create a physical
network-layer datagram
source • Hn used by network layer protocol to destination
implement its service
Introduction: 1-32
Services, Layering and Encapsulation
M

application application
Ht M
transport transport
Hn Ht M
network Network-layer protocol transfers transport-layer segment
network
[Ht | M] from one host to another, using link layer services
Hl Hn Ht
link M
link
Link-layer protocol transfers datagram [Hn| [Ht |M] from
host to neighboring host, using network-layer services
physical  link-layer protocol encapsulates network physical
datagram [Hn| [Ht |M], with link-layer header
source Hl to create a link-layer frame destination

Introduction: 1-33
Services, Layering and Encapsulation
M

application M application
message
Ht M
transport Ht M
transport
segment
Hn Ht M Hn Ht M
network network
datagram

Hl Hn Ht Hl Hn Ht M
link M
link
frame

physical physical

source destination

Introduction: 1-34
source Encapsulation:
message
segment Htt
M
M
application
transport
an end-end
datagram Hn Ht M network view
frame Hl Hn Ht M link
physical
link
physical

switch

destination Hn Ht M network
M application Hl Hn Ht M link Hn Ht M
Ht M transport physical
Hn Ht M network
Hl Hn Ht M link router
physical
Introduction: 1-35
Lecture 1: roadmap
 What is the Internet?
 What is a protocol?
 Network edge: hosts, access
network, physical media
 Network core: packet/circuit
switching, internet structure
 Performance: loss, delay,
throughput
 Security
 Protocol layers, service models
Introduction: 1-36
Chapter 1: summary
We’ve covered a “ton” of material!
 Internet overview
 what’s a protocol? You now have:
 network edge, access network, core  context, overview,
• packet-switching versus circuit-
switching vocabulary, “feel”
• Internet structure of networking
 performance: loss, delay, throughput  more depth,
 layering, service models detail, and fun to
 security follow!

Introduction: 1-37

You might also like