Introduction to Computer Networks

Network Essentials Session 1 R. Venkateswaran

Outline

Session 1

7-Layer OSI Model

Network Layer protocols (Internet Protocol)

Transport Layer protocols (TCP and UDP)

Session 2

Socket Programming – with focus on BSD Sockets

Sample codes in C that work on UNIX/Linux systems

Acknowledgments

This presentation has been adapted from presentations available at:

Prof. Shivkumar Kalyanaraman (http://www.ecse.rpi.edu/Homepages/shivkuma/)

Prof. Sneha Kumar Kasera (http://www.cs.utah.edu/classes/cs5480/)

Prof. David Hollinger (http://www.cs.rpi.edu/~hollingd/netprog)

South Asian Network Operators Group (http://ws.edu.isoc.org/workshops/2004/SANOG-IV/ip-services/presentations/ip-intro/ipbasics.ppt)

Network Models

Formal models allow us to deal with various aspects of networks abstractly

One such model is the OSI reference model

The OSI reference model is a layered model

Divide a task into pieces and then solve each piece independently

Establishing a well defined interface between layers

Major Advantages:

Each layer can be implemented independently

Adaptability

Code Reuse

Extensibility

OSI 7-Layer Model IP TCP/UDP Virtual End-to-end connectivity Path selection, Internetworking Error-free communication links Transmission of raw signal Mail, Web, etc. 1 3 2 4 5 6 7 Ethernet Data encryption,compression Managing sessions Funtionality Examples Layers Application Presentation Session Transport Network Data Link Physical

OSI 7-Layer Host Router Router Host End to end Hop by hop Application Presentation Session Transport Network Link Physical Network Link Link Network Link Link Physical Physical Application Presentation Session Transport Network Link

TCP/IP Model Host Router Router End to end No session or presentation layers in TCP/IP model Host Hop by hop Application Transport Network Link Physical Network Link Link Network Link Link Physical Physical Application Transport Network Link

Packet structure Trailer Header Header Header Application Transport Network Data Link Data Transport Layer Data Network Layer Data Link Layer Data

Network Layer

Internet Architecture

Packet-switched, connectionless datagram network

IP is the network layer protocol

Acts as a glue

Hourglass concept

all hosts and routers run IP

Stateless architecture

no per flow state inside the network

Hop-by-hop packet forwarding

Header contains all the information

IP - Minimalist Approach

Dumb network

Connectivity is the key

Network provides minimal functionalities to support connectivity

Addressing, forwarding, routing

Smart end systems

Transport layer or application performs more sophisticated functionalities

Flow control, error control, congestion control

Advantages

High scalability

Works across heterogeneous technologies (Ethernet, modem, satellite, wireless)

Supports diverse applications (telnet, ftp, Web, media streaming)

Decentralized network administration

IPv4 Header Transport Layer Data… 0 4 8 16 32 IHL Type of Service Total Length Version Fragment Offset Identification Flags Time to Live Protocol Header Checksum Source Address Destination Address Padding Options

IP Address

IP address: Unique identification of the end-system from a network-layer perspective

IP address is 32-bits long (version 4)

Contains a network ID and host ID

Use subnet mask to detect the network ID

Example of IP address:

133.27.162.125

133 27 162 125 10000101 00011011 10100010 01111101 85 1B A2 7D Decimal Binary HEX NetID Host ID Boundary

Network Mask

Define which bits are used to describe the network ID

Different Representations:

Decimal dot notation: 255.255.224.0

Number of network bits: /19

Bitwise-AND of 32-bit IP address with 32-bit netmask yields network ID part of the address (truncated appropriately)

Network Mask Examples 137.158.128.0/ 17 (netmask 255.255.128.0 ) 1111 1111 1111 1111 1 000 0000 0000 0000 1111 1111 1111 1111 0000 0000 0000 0000 1111 1111 1111 1111 1111 1111 11 00 0000 198.134.0.0/ 16 (netmask 255.255.0.0 ) 205.37.193.128/ 26 (netmask 255.255.255.192 ) 1000 1001 1001 1110 1 000 0000 0000 0000 1100 0110 1000 0110 0000 0000 0000 0000 1100 1101 0010 0101 1100 0001 10 00 0000

Subnets

All device interfaces having the same network ID are part of the same subnet

Devices within a subnet can communicate with each other without an intervening router

Network consisting of 3 subnets 192.1.1.2 192.1.1.1 192.1.1.3 192.1.1.4 192.1.2.9 192.1.2.2 192.1.2.1 192.1.3.2 192.1.3.1 192.1.3.27

IP router

A device with more than one link-layer interface

Each interface identified by a different IP address (from different subnets)

Packets arriving at one interface are forwarded out on another interface to get them closer to the destination

Creates and maintains forwarding tables

Tables help in making forwarding decisions

Tables created and updated based on routing information exchanged between routers

Each router maintains its own forwarding table

Hop by Hop Forwarding

IP Forwarding Rules - I

Destination is in the same subnet (direct connectivity)

Recognize that destination IP address is on same subnet

Find the destination’s datalink-layer address

IP packet encapsulated and sent directly to the destination’s datalink-layer address

Destination is in a different subnet (indirect connectivity)

Recognize that destination IP address is on different subnet

Look up destination IP address in a (L3 forwarding) table to find a match, called the next hop router IP address

Find the next hop router’s datalink-layer address

IP packet encapsulated and sent directly to the next hop router’s datalink-layer address

IP Forwarding Rules - II

Problem 1: Recognize if destination is on the same subnet

Use netmask to compute network ID of the destination and match it with device’s network ID

Problem 2: Find a device’s datalink-layer address

Static mapping

Dynamic mapping using Address Resolution Protocol (ARP)

Sender host broadcasts a request: “ What is the Ethernet address of 192.1.1.4? ”

The device whose IP address is 192.1.1.4 replies back: “ The Ethernet address for 192.1.1.4 is 00-0C-F1-4E-2A-E2 ”

ARP responses are cached at the sender

Use arp command to view/modify the cache

Look up Forwarding Table Destination =12.5.9.16 ------------------------------- payload Prefix Interface Next Hop 12.0.0.0/8 10.14.22.19 eth1 12.4.0.0/15 12.5.9.0/24 attached eth2 Serial 1/0/7 10.1.3.77 IP Forwarding Table 0.0.0.0/0 10.14.11.33 eth0 even better OK better best! Longest Prefix Match (Classless) Forwarding

IP Forwarding – Example 1 Forwarding Table on host 192.1.1.1 Note: 127.0.0.1 is the special address of the local interface 192.1.3.2 192.1.1.2 192.1.1.1 192.1.1.3 192.1.1.4 192.1.3.1 192.1.3.27

Routing Protocols

Distance Vector Routing protocol

Each router sends a vector of distances to its neighbors

Vector contains distances to all the nodes

Each router computes next hop towards different nodes

Iterative, Asynchronous, Distributed computation

Link State Routing Protocol

Each router sends a vector of distances to all the nodes

Vector contains distances to only the neighbors

Each router has complete topology information

Can compute shortest paths to different nodes

Distance vector algorithm - I

Basic idea:

Each node periodically sends its own distance vector estimate to neighbors

When a node x receives new DV estimate from neighbor, it updates its own DV using Bellman-Ford equation:

D x (z) ← min v {c(x,v) + D v (z)} for each node z ∊ N Nexthop x (z) = v

Under natural conditions, the estimate D x (z) converges to the actual least cost d x (z)

2 0 1 from from x y z x y z 0 2 3 from cost to x y z x y z 0 2 3 from cost to x y z x y z ∞ ∞ ∞ ∞ ∞ cost to x y z x y z 0 2 7 from cost to x y z x y z 0 2 3 from cost to x y z x y z 0 2 3 from cost to x y z x y z 0 2 7 from cost to x y z x y z ∞ ∞ ∞ 7 1 0 cost to ∞ ∞ ∞ 2 0 1 7 1 0 2 0 1 7 1 0 2 0 1 3 1 0 2 0 1 3 1 0 2 0 1 3 1 0 2 0 1 3 1 0 time node x table node y table node z table D x (y) = min{c(x,y) + D y (y), c(x,z) + D z (y)} = min{2+0 , 7+1} = 2 D x (z) = min{ c(x,y) + D y (z), c(x,z) + D z (z) } = min{2+1 , 7+0} = 3 ∞ x y z x y z 0 2 7 ∞ ∞ ∞ ∞ ∞ ∞ from cost to x z 1 2 7 y

Distance Vector: link cost changes

Link cost changes:

node detects local link cost change

updates routing info, recalculates distance vector

if DV changes, notify neighbors

“ good news travels fast” At time t 0 , y detects the link-cost change, updates its DV, and informs its neighbors. At time t 1 , z receives the update from y and updates its table. It computes a new least cost to x and sends its neighbors its DV. At time t 2 , y receives z ’s update and updates its distance table. y ’s least costs do not change and hence y does not send any message to z . x z 1 4 50 y 1

Distance Vector: link cost changes

Link cost changes:

good news travels fast

bad news travels slow - “count to infinity” problem!

44 iterations before algorithm stabilizes: see text

Poisoned reverse:

If Z routes through Y to get to X :

Z tells Y its (Z’s) distance to X is infinite (so Y won’t route to X via Z)

will this completely solve count to infinity problem?

x z 1 4 50 y 60

Comparison of LS and DV algorithms

Message complexity

LS: with n nodes, E links, O(nE) msgs sent

DV: exchange between neighbors only

convergence time varies

Speed of Convergence

LS: O(n 2 ) algorithm requires O(nE) msgs

may have oscillations

DV : convergence time varies

may be routing loops

count-to-infinity problem

Robustness: what happens if router malfunctions?

LS:

node can receive incorrect link cost

each node computes only its own table

DV:

DV node can advertise incorrect path cost

each node’s table used by others

error propagate thru network

Transport Layer

Transport Protocols

Protocol implemented entirely at the ends

Completeness/correctness of function implementations

User Datagram Protocol (UDP) provides just integrity and demultiplexing

Transmission Control Protocol (TCP) adds…

Connection-orientedness (point-to-point)

Reliability

In-Order delivery

Byte-stream

Full duplex

Flow and congestion control

UDP: User Datagram Protocol [RFC 768]

“ No frills”, “bare bones” Internet transport protocol

“ best effort” service, UDP segments may be:

lost

delivered out of order to the application

connectionless:

no handshaking between UDP sender, receiver

each UDP datagram handled independently of others

Why is there a UDP?

No connection establishment => Faster communication

simple: no connection state at sender, receiver

small 8-byte header (lower overheads)

no congestion control: UDP can blast away as fast as desired

Streaming Multimedia apps, DNS, SNMP benefit from UDP

0 16 32 Source Port Destination Port Length Checksum Application Layer Data…

TCP Header 0 Source Port Destination Port Sequence Number Acknowledgement Number Data Offset Window Reserved ACK URG EOL RST SYN FIN Checksum Urgent Pointer Padding Options 4 8 16 32 Application Layer Data…

Connection: Three-Way Handshake

TCP connection-establishment: 3-way-handshake necessary and sufficient for unambiguous setup/teardown even under conditions of loss, duplication, and delay

TCP Connection Setup: FSM CLIENT SERVER

TCP – Streams-based Host A Seq=100, 20 bytes data ACK=100 Host B Seq=92, 8 bytes data ACK=120 SendBase = 120 Sendbase = 100 time

TCP is Network-friendly

Reliable transmission

Sliding window concept

Flow control

Regulated by the receiver

Congestion Control

Regulated by the sender

Additive Increase, Multiplicative Decrease

Fairness of TCP streams

Stop-and-Wait operation first packet bit transmitted, t = 0 sender receiver RTT last packet bit transmitted, t = L / R first packet bit arrives last packet bit arrives, send ACK ACK arrives, send next packet, t = RTT + L / R

Pipelining: increased utilization first packet bit transmitted, t = 0 sender receiver RTT last bit transmitted, t = L / R first packet bit arrives last packet bit arrives, send ACK ACK arrives, send next packet, t = RTT + L / R last bit of 2 nd packet arrives, send ACK last bit of 3 rd packet arrives, send ACK Increase utilization by a factor of 3!

Go-Back-N

Sender:

k-bit seq # in pkt header

“ window” of up to N, consecutive unack’ed pkts allowed

ACK(n): ACKs all pkts up to, including seq # n - “cumulative ACK”

may receive duplicate ACKs (see receiver)

timer for each in-flight pkt

timeout(n): retransmit pkt n and all higher seq # pkts in window

Go-Back-N

TCP Flow Control

receive side of TCP connection has a receive buffer:

speed-matching service: matching the send rate to the receiving app’s drain rate

app process may be slow at reading from buffer

sender won’t overflow receiver’s buffer by transmitting too much, too fast flow control

TCP Flow control: how it works

(Suppose TCP receiver discards out-of-order segments)

spare room in buffer

= RcvWindow

= RcvBuffer-[LastByteRcvd - LastByteRead]

Rcvr advertises spare room by including value of RcvWindow in segments

Sender limits unACKed data to RcvWindow

guarantees receive buffer doesn’t overflow

TCP congestion control:

two “phases”

slow start

congestion avoidance

important variables:

Congwin

threshold: defines threshold between two slow start phase, congestion control phase

“ probing” for usable bandwidth:

ideally: transmit as fast as possible ( Congwin as large as possible) without loss

increase Congwin until loss (congestion)

loss: decrease Congwin , then begin probing (increasing) again

TCP Slowstart

exponential increase (per RTT) in window size (not so slow!)

loss event: timeout (Tahoe TCP)

initialize: Congwin = 1 for (each segment ACKed) Congwin++ until (loss event OR CongWin > threshold) Host A one segment RTT Host B two segments four segments Slowstart algorithm time

TCP Congestion Avoidance: Tahoe /* slowstart is over */ /* Congwin > threshold */ Until (loss event) { every w segments ACKed: Congwin++ } threshold = Congwin/2 Congwin = 1 perform slowstart TCP Tahoe Congestion avoidance

Where to from here?

IP designed for best-effort service only

Affecting new applications like media streaming, VoIP

Should the network become application-aware?

Should the IP routers look beyond the IP header?

We are already running short of IP addresses

Solution: IPv6 – yet to become widespread

Temporary fix: Network Address Translation (NAT)

TCP or UDP may not be the best suited for reliable media streaming

Answer: Stream Control Transmission Protocol (SCTP)

SCTP combines the datagram orientation of UDP with the sequencing and reliability of TCP

SCTP uses multi-streaming, message-oriented routing

Outline

Session 1

7-Layer OSI Model

Network Layer protocols (Internet Protocol)

Transport Layer protocols (TCP and UDP)

Session 2

Socket Programming – with focus on BSD Sockets

Sample codes in C that work on UNIX/Linux systems

Thank You !

BACKUP

TCP retransmission - I

Premature Timeout

Host A Seq=100, 20 bytes data ACK=100 time Host B Seq=92, 8 bytes data ACK=120 Seq=92, 8 bytes data Seq=92 timeout ACK=120 Seq=92 timeout SendBase = 120 SendBase = 120 Sendbase = 100

TCP retransmission - II

Lost ACK packet

Host A Seq=92, 8 bytes data ACK=100 loss timeout Host B X Seq=92, 8 bytes data ACK=100 time SendBase = 100

TCP Connection Tear-down Sender Receiver FIN FIN-ACK FIN FIN-ACK Data write Data ack

TCP Connection Tear-down: FSM CLOSING CLOSE WAIT FIN WAIT-1 ESTAB TIME WAIT snd FIN CLOSE send FIN CLOSE rcv ACK of FIN LAST-ACK CLOSED FIN WAIT-2 snd ACK rcv FIN delete TCB Timeout=2msl send FIN CLOSE send ACK rcv FIN snd ACK rcv FIN rcv ACK of FIN snd ACK rcv FIN+ACK

TCP retransmission - III