INTERNETWORKING
Introduction and Overview
Motivation for Internetworking
Internetworking – technology forms the basis for the internet by accommodating multiple diverse underlying hardware technologies , providing a way to interconnect the networks, and defining a set of communication conventions that the network used to interoperate
The TCP/IP Internet
• DARPA (Defence Advanced Research Projects Agency)- It includes a set of standards that specify the details of how computers communicate
- a set of conventions for interconnecting networks and forwarding traffic
- Officially named TCP/IP Internet Protocol Suite and commonly referred to as TCP/IP
- TCP/IP can be used to interconnect a set of networks within a single building, within a physical campus or among a set of campuses
- TCP/IP forms the base technology for the global internet that connects over 650 million individuals in homes, schools, corporations and government labs in virtually all populated areas
Internet Services
• Focus on standards called protocols . Protocols like TCP and IP provide the syntactic and semantic rules for communication.
- They contain details of the message formats, describe how a computer responds when a message arrives and specify how a computer handles errors or other abnormal conditions.
- communication protocol allows to specify or understand data communication independent on detailed knowledge of any particular vendor’s network hardware
- Hiding the low-level details of communication helps improve productivity in several ways.
1. Programmers deal with higher-level protocol abstractions, they do not need to learn or remember as many details about a given hardware configuration. Thus, they create new programs quickly
2. Since programs are not restricted to a particular computer architecture or a particular network hardware, the programs need not be changed when computers or networks are replaced or reconfigured
3. As application programs built using higher-level protocols independent of underlying hardware, they can provide direct communication between an arbitrary pair of computers and no need to build an application software for each type of computer or each type of network.
Application Level Internet Services
• Internet application programs exhibit a high degree of interoperability (refers to the ability of diverse computing systems to cooperate in solving computational problems).
• World wide web(www)- The web allows users to view documents that contain text and graphics and to follow hypermedia links from one document to another
• E-Mail- Allows user to compose a message and send a copy to individuals or groups and also read memos that they have received. One reason for popularity is : protocol makes delivery reliable
• File Transfer- The file transfer application allows users to send or receive a copy of a data file.
• Remote Login and Remote Desktop- allow users sitting at one computer to connect to a remote machine as local
ie keystrokes are sent to the remote machine and display from the remote machine appears on the user’s screen (either in a window or across the entire screen)
Network-Level Internet Services
• At the network level, an internet provides 2 broad types of services that all application programs use.
• Connectionless Packet Delivery Service: TCP/IP internet forwards small messages from one computer to another based on address information carried in the message.
- It forwards each packet separately , so does not guarantee reliable, in-order delivery
- It maps directly onto the underlying hardware, the connectionless service is extremely efficient.
• Reliable Stream Transport Service:
- It allows an application on one computer to establish a “connection” with an application on another computer and then to send a large volumes of data across the connection as it were a permanent, direct connection.
- The underneath communication protocol divide the stream of data into small messages and send them one at a time, waiting for the receiver to acknowledge reception
- It handles problems like recover automatically from transmission errors, lost packets or failures of intermediate switches along the path between the sender and receiver
The features of TCP/IP
- Network Technology Independence: independent of any particular brand or type of hardware . TCP/IP protocol define unit of data transmission called a datagram and specify how to transmit datagrams on a particular network
- Universal Interconnection- allows any pair of computers to communicate. Each computer is assigned an address that is universally recognized throughout the network. Each datagram carries the address of the source and destination
- End-to-End Acknowledgements: provide acknowledgements between the original source and ultimate destination instead of successive machines along the path.
- Application Protocol Standards: In addition to the basic transport-level services like reliable stream connections), the TCP/IP protocols include standards for common applications including e-mail, file transfer and remote login
.
Network Technologies
Two Approaches to Network Communication
1. Connection oriented(sometimes called circuit switched)
2. Connectionless(sometimes called packet switched)
Connection oriented networks
It operates by forming a dedicated connection or circuit between two points eg. Traditional telephone system
Adv: Guaranteed capacity of the network(ie. Once a circuit is established, no other network activity decrease the capacity of the network)
Disadv: Cost(circuit costs are fixed and independent of use)
Connectionless networks
• Data to be transferred across the network is divided into small pieces called packets and are sent at a time.
• A packet contains few hundred bytes of data, carries identification that enables the network h/w to know how to send to destination
• The network h/w delivers to the destination and the s/w reassembles them into a single file again
Adv
1. Multiple communication among computers can proceed concurrently with intermachine connections shared by all pairs of computers that are communicating
2. low cost and performance(since multiple computers shares network bandwidth, fewer connections are required)
Disadv: less network capacity (ie, as activity increases, network becomes overloaded, computers should wait before they send additional packets
Packet switched networks are divided into
1. WAN 2. LAN
WAN(called long haul networks):
1. provide communication over long distances
2. Do not limit the distance spanned
3. operates at slower speeds(1.5Million bits per sec to 2.4 billion bits per sec) than LANs and have much greater delay( from few milliseconds to several tenths of a second) between connections
4. a network contains a series of complex computers called packet switches interconnected by long-distance communication lines.
5. The size of the network can be extended by adding a new switch and another communication line. Attaching a user’s computer to a WAN means connecting it to one of the packet switches
LAN
• Provide high speed connections among the computers and operates between 100Mbps and 10Gbps
• Spans a small area like single building or a single campus. So, lower delays( as short as few tenths of a millisecond or as long as 10 milliseconds) than WANs
• Each computer contains a device known as Network Interface Card(NIC) that connects the machine directly to the network.
Network Hardware Addresses
• Each hardware technology defines an addressing mechanism that computers use to specify the destination for a packet
• Every computer connected to network is assigned an unique address usually integer
• Every packet contains destination address field that appears in the same location in all packets
• Each hardware technology specifies how computers are assigned addresses
eg. h/w specifies the no: of bits in the address as well as the location of the destination address field in the packet.
Ethernet Technology
• Ethernet is the name given to popular packet-switched LAN technology invented at Xerox PARC in the early 1970s.
• Xerox Corporation, Intel Corporation and Digital Equipment Corporation standardized ethernet in 1978.
• IEEE released a compatible version of the standard using the standard number 802.3
• Ethernet has become the most popular LAN technology and the original wiring is coaxial cable has been out-of-date
• The current technology is twisted pair Ethernet and it allows computers to access ethernet using conventional unshielded copper wires(similar to wires used to connect telephones)
Adv of twisted pair wiring (category 5 cable) is, it reduces costs and is easier to install than coaxial cable
• The first twisted pair Ethernet known as 10Base-T operated at 10Mbps exactly like orginal ethernet
• A set of 8 wires(4 pairs) is used to connect each computer to a central Ethernet hub or switch
• Only 4 of 8 wires are used : one pair carries data from the computer to the hub and another pair carries data from the hub to the computer
• Hub is an electronic device and it is small box that resides in a wiring closet. A connection between a hub and a computer must be less than 100m long. Hub requires power and allow authorized personnel to monitor and control its operation over the network.
Ethernet capacity
• The initial twisted pair Ethernet design operated at the same rate as the original Ethernet : data could be transmitted at 10 million bits per second
• High bandwidth(amount of data that can be carried from one point to another within a given period of time) carries heavy traffic loads
• In the late 1970s when ethernet was standardized , a LAN operating at 10 MBps has more sufficient capacity
• For more CPU speeds , ethernet operating at 10Mbps and so ethernet becomes bottleneck
ATM
• Name given to connection oriented networking for both LAN and WAN
• Designed to permit high-speed data switching
and it is achieved thru special-purpose h/w and s/w
• ATM network consists of one or more high speed switches that each connect to computers and to other switches
• The lowest layers of ATM network uses fixed size frames called cells
• Each cell is exactly of fixed size and ATM switch h/w can process cells quickly
• Each ATM cell is only 53 octets long(5 octets of header followed by 48 octets of data
Connection Oriented Networking
• First computer interacts with the switch Connection (telephone call) to specify the destination and waits until switch finds the path thru the n/w and establishes the connection
• Once the connection succeeds , ATM switch chooses an identifier for the connection and passes it to the computer for sending/recieing cells
• When it finishes using an ATM connection, a computer again communicates with the ATM switch to break the connection
• The switch then disconnects the two computers(like hanging up the telephone call)
• Identifiers used for connection can be recycled for the another connection
Wide area point-to-point networks
From TCP/IP point of view, any communication
system that connects exactly two computer is known as point-to-point network.
eg Dialup IP- Dialup internet access
dialup modem in a computer is used to place a phone call to another modem(two modem use audio tones to send data)
Internetworking Concept and Architectural Model
Application-Level Interconnection
• Two approaches to hide network details
1. using application programs to handle heterogenity (application-Level Interconnection)
2. Operating system (application-Level Interconnection)
• Each heterogeneous network interconnections provided uniformity through application programs called application gateways
• Application programs executing on each computer understands the details of the network connections for that computer, and interoperates across those connections with application programs on other computers
eg Electronic mail systems
Disadv: 1. Limited and cumbersome communication (If n/w grows to 100s or 1000s n/ws , no one can build all the necessary application programs)
2. No guaranteed reliable communication( intermediate program fails, both source and destination unable to detect or control the problem)
Network-Level Interconnection
• It provides the mechanism to deliver packets of data from their original source to ultimate destination without using intermediate application programs.
Switching small packets has many advantages
1. It maps directly onto the underlying n/w hardware making it extremely efficient
2. n/w level interconnection separates data communication activities from application programs, permitting intermediate computers to handle traffic without understanding the applications that are sending or receiving.
3. Using n/w connections keeps the entire system flexible (making possible to build general purpose communication facilities)
4. Allows Network managers to add new technologies by modifying or adding a single piece of new network level software while application programs remains unchanged.
The key to design universal network-level interconnections can be found in an abstract communication system concept known as internetworking
Properties of the Internet
All computers in the internet share a universal set of machine identifiers as names or addresses
Users or application programs need not want to understand the details of hardware interconnections to use the internet
Internet Architecture
How are networks interconnected to form an internetwork?
Internet Gateways or Internet routers are computers that interconnect two networks and pass packets from one to the other
Interconnection Through IP routers
In actual internet containing many networks and routers, each router need to know the topology of the internet beyond the networks to which it connects.
Ex. 3 networks interconnected by two routers(R1 sends packets to either n/w1 or N/w2
Routers use the destination network, not the destination computer when forwarding a packet.
Routers have little disk storage and modest main memories
The user’s view
• The users view an internet as a single, virtual network to which all machines connect despite of their physical connections
- It makes easy to conceptualize communication
- In addition to router that interconnect physical networks, software is needed on each computer to allow application programs to use an internet as if it were a single, physical network
Classful Internet Address
Universal Identifiers: Globally accepted method of identifying each computer that attaches to it.
- The host identifiers are classified as names, addresses or routes
- software takes the identifiers as addresses
Original classfull addressing scheme
• Each host on the internet is assigned a 32-bit integer address called its Internet Protocol address or IP address
• IP address encodes the identification of the network to which host attaches as well as the identification of a unique host on that network.
• Each address is a pair (netid, hostid)
netid- identifies a network
hostid – identifies the host on that network
• The original addressing scheme which is known as classful, each IP address had one of the first three forms
1.Class A- used for handful of n/ws that have
more than 65536 hosts, devote 7 bits to netid
and 24bits to hostid
2. Class B addreses-used for intermediate size networks that have between 256 to 65536,
- Allocate 14 bits to the netid and 16 bits to the hostid
3.Class c- used for n/ws that have less than 256 hosts, devote 21 bits to netid and 8 bits to hostid
4. Class D
5. Class E
Network and directed broadcast Addresses
Adv of IP address:
1. Makes efficient forwarding possible
2. Internet addresses can be used to refer to networks as well as individual hosts.
By convention, an address that has all bits of the hostid is equal to 0 is reserved to refer to the network
3. Direct Broadcast Address: any address with the hostid all 1s is reserved for direct broadcast address.
- It refers to all hosts in the network(The packet is sent to all computers on a network)
Disadv: It requires knowledge of the network address.
Limited broadcast
• Another broadcast address is called the local broadcast address.
• It consists of 32 1’s(hence sometimes called “all 1s” )
• It provides the broadcast address for the local network independent of the assigned IP address.
• A host can use the limited broadcast as a start up procedure before it learns its IP address. Once the host learns the correct IP address for the local network, however directed broadcast is preferred
The ALL-0s Address
• An address consists of 32 zero bits is reserved for cases where a host needs to communicate but does not yet knows its IP address
• A host sends a datagram to the limited broadcast address and uses address 0 to identify itself.
• The receiver understands that the host doest not yet have an IP address and the receiver uses a special method to send a reply
IP Multicast Addresses
Unicast delivery: packet is delivered to a single computer
Broadcast delivery: packet is delivered to all computers on a given network.
Multicasting: packet is delivered to a specific subset of hosts.
Weaknesses in Internet Addressing
If a host moves from one network to another network, its IP address must change
Once a prefix is chosen, max no: of hosts on the network is fixed. If the n/w grows beyond the original bound, a new prefix must be selected and all the hosts must be renumbered (that are time consuming and difficult to debug)
Internet addressing scheme is not fully apparent until forwarding(b’cas forwarding uses the netid of the IP address, the path taken by packets traveling to a host with multiple IP address )
Dotted Decimal Notation
IP address written as four decimal integers separated by decimal points
Each Integer gives the value of one octet of the IP address(thus 32bit IP address)
eg 10000000 00001010 00000010 00011110
is written as 128.10.2.30
Range of dotted decimal values
Class Lowest address Highest address
A 1.0.0.0 127.0.0.0
B 128.0.0.0 191.255.0.0
C 192.0.0.0 223.255.255.0
D 224.0.0.0 239.255.255.255
C 240.0.0.0 255.255.255.254
Loopback address
• The network prefix 127.0.0.0 – reserved for loopback
• It is intended for use in testing TCP/IP and for interprocess communication on the local computer
• When program uses loopback address as a destination, the protocol s/w processes data without sending traffic across any network.
• A datagram sent to network 127 address should never appear on any network
• Host or router should never propagate routing/ reachability information for network number 127
Reserved address prefix
• For private intranet(not connected to the Internet)
Eg class A address 9.0.0.0 assigned to IBM
address 12.0.0.0 assigned to AT&T
• To avoid conflict between private internet and global Internet,IETF reserved several address prefixes and recommends them to use on private internets
• Each network will have network numbers like 9.0.0.0
• Each host connected to the network (i.e) each network connection is assigned an IP address
Network Byte Order
Eg passing 32-bit integers from one host to another (not all architectures store 32-bit in the same way)
Little Endian- lowest memory address contains low-order byte of the integer
Big Endian- lowest memory address contains high-order byte of the integer
• Network standard byte order- It defines that all machines must use for binary fields in internet packets.
• Each router or host converts binary items to network standard byte order before sending a packet and converts from network byte order to host specific order when a packet arrives
• The internet standard for byte order specifies – Big Endian Style
Mapping Internet Addresses to Physical Addresses(ARP)
The address resolution problem
• Mapping must be done in the path from the source to the destination. Source sending data should map the Internet address to the physical address of the destination/intermediate router
• The problem of mapping high-level addresses to physical addresses is known as address resolution problem(solved by maintaining tables or encoding hardware addresses in high-level addresses
Two types of addresses
• Large, fixed addresses like ethernet
• Small, easily configured addresses
Resolution thru Direct mapping
If the IP address encodes the hardware address, the hardware address can be extracted from the low-order octet
Adv:
1. Computationally efficient( it requires few machine instructions)
2. Does not involve references to external data
3. New computers can be added without changing existing assignments or propagating information to existing computers.
Disadv:
It cannot be used for hardware technologies that use Ethernet Addressing(each ethernet NIC card Is assigned a 48-bit physical address when the device is manufactured)
Resolution thru Dynamic binding
Address Resolution Protocol provides the mechanism to bind the addresses dynamically that is both reasonably efficient and easy to maintain
Suppose host A wants to communicate to host B
- Host A broadcasts a ARP request (containing internet address of B) that asks the host B to respond with its physical address
- All hosts including B receives the request but B only recognizes its IP address and sends a ARP reply that contains its physical address
ARP Cache Timeout
• ARP Cache is used
- To reduce the communication costs
- It maintain recently acquired IP-to-physical address bindings.( whenever a computer sends ARP packet and receives reply, it saves the IP address and corresponding hardware address)
- Before transmitting a packet , a host looks in its cache for a binding. If it finds, host need not broadcast
• ARP cache provides “soft state” – in which information can become “stale” without warning
eg. From host A to B communication – after information exchange, B crashes, but the address binding is in cache and A sends information to B using the “stale” address binding in cache.
Host A has no way of knowing when information in its ARP cache is incorrect
• To accommodate the soft state, a timer has been set whenever address binding is placed in an ARP, it will be timeout after some period of time (eg 20 minutes)
The use of soft state has advantages and disadvantages
Adv
1. A computer can determine when information in its cache can be revalidated independent of other computers
2. Sender does not need successful communication with a receiver or the third party to determine that the binding is invalid
Disadv
There is a delay if the receiver has crashed before the time interval of the timer expires and the sender does not detect it.
ARP Refinements
• The sender’s IP-to-physical address binding is included in every ARP broadcast; receivers update the IP-to-physical address binding information in their cache before processing an ARP packet( to avoid extra traffic)
• When a computer has its host interface replaced, its physical address changes. The computer notify others of a new address by broadcasting a gratuitous ARP request when it boost.
0 comments:
Post a Comment