TCP/IP Protocol Architecture

TCP/IP defines a large collection of protocols that allow computers to communicate. Table 1.1 outlines the protocols and the TCP/IP architectural layer to which they belong. TCP/IP defines the details of each of these protocols in Requests For Comments (RFC) documents. By implementing the required protocols defined in TCP/IP RFCs, a computer that implements the standard networking protocols defined by TCP/IP can communicate with other computers that also use the TCP/IP standards.

{ 0 comments }

TCP/IP and the OSI Reference Model

TCP/IP model consists of four layers, each of which can have several sublayers. These layers correlate roughly to layers in the OSI reference model and define similar functions. Some of the TCP/IP layers correspond directly with layers in the OSI reference model while other span several OSI layers.

• The TCP/IP Application Layer refers to communications services to applications and is the interface between the network and the application. It is also responsible forpresentation and controlling communication sessions. It spans the Application Layer, Presentation Layer and Session Layer of the OSI reference model. Examples include: HTTP, POP3, and SNMP.

• The TCP/IP Transport Layer defines several functions, including the choice of protocols, error recovery and flow control. The transport layer may provide for retransmission, i.e., error recovery, and may use flow control to prevent unnecessary congestion by attempting to send data at a rate that the network can accommodate, or it might not, depending on the choice of protocols. Multiplexing of incoming data for different flows to applications on the same host is also performed. Reordering of the incoming data stream when packets arrive out of order is included. It correlates with the Transport Layer of the OSI reference model. Examples include: TCP and UDP, which are called Transport Layer, or Layer 4, protocols.

• The TCP/IP Internetwork Layer defines end-to-end delivery of packets and defines logical addressing to accomplish this. It also defines how routing works and how routes are learned; and how to fragment a packet into smaller packets to accommodate media with smaller maximum transmission unit sizes. It correlates with the Network Layer of the OSI reference model. Examples include: IP and ICMP

• The TCP/IP Network Interface Layer is concerned with the physical characteristics of the transmission medium as well as getting data across one particular link or medium. This layer defines delivery across an individual link as well as the physical layer pecifications. It spans the Data Link Layer and Physical Layer of the OSI reference model. Examples include: Ethernet and Frame Relay.

{ 0 comments }

Interaction Between OSI Layers

When a host receives a data transmission from another host on the network, that data is processed at each of the OSI layers to the next higher layer, in order to render the data transmission useful to the end-user. To facilitate this processing, headers and trailers are created by the sending host’s software or hardware, that are placed before or after the data given to the next higher layer. Thus, each layer has a header and trailer, typically in each data packet that comprises the data flow. The sequence of processing at each OSI layer, i.e., the processing between adjacent OSUI layers, is as follows:

• The Physical Layer (Layer 1) ensures bit synchronization and places the received binary pattern into a buffer. It notifies the Data Link Layer (Layer 2) that a frame has been received after decoding the
incoming signal into a bit stream. Thus, Layer 1 provides delivery of a stream of bits across the medium. [click to continue…]

{ 0 comments }

OSI Reference Model

The OSI Reference Model

The OSI is the Open System Interconnection reference model for communications. the OSI reference model consists of seven layers, each of which can have several sublayers. The upper layers of the OSI reference model define functions focused on the application, while the lower three layers define functions focused on end-to-end delivery of the data.

• The Application Layer (Layer 7) refers to communications services to applications and is the interface between the network and the application. Examples include: Telnet, HTTP, FTP, Internet browsers, NFS, SMTP gateways, SNMP, X.400 mail, and FTAM.

• The Presentation Layer (Layer 6) defining data formats, such as ASCII text, EBCDIC text, binary, BCD, and JPEG. Encryption also is defined as a presentation layer service. Examples include: JPEG, ASCII, EBCDIC, TIFF, GIF, PICT, encryption, MPEG, and

• The Session Layer (Layer 5) defines how to start, control, and end communication sessions. This includes the control and management of multiple bidirectional messages so that the application can be notified if only some of a series of messages are completed. This allows the presentation layer to have a seamless view of an incoming stream of data. The presentation layer can be presented with data if all flows occur in some cases. Examples include: RPC, SQL, NFS,
NetBios names, AppleTalk ASP, and DECnet SCP

• The Transport Layer (Layer 4) defines several functions, including the choice of protocols. The most important Layer 4 functions are error recovery and flow control. The transport layer may provide for retransmission, i.e., error recovery, and may use flow control to prevent unnecessary congestion by attempting to send data at a rate that the network can accommodate, or it might not, depending on the choice of protocols. Multiplexing of incoming data for different flows to applications on the same host is also performed. Reordering of the incoming data stream when packets arrive out of order is included. Examples include: TCP, UDP, and SPX.

• The Network Layer (Layer 3) defines end-to-end delivery of packets and defines logical addressing to accomplish this. It also defines how routing works and how routes are learned; and how to fragment a
packet into smaller packets to accommodate media with smaller maximum transmission unit sizes. Examples include: IP, IPX, AppleTalk DDP, and ICMP. Both IP and IPX define logical addressing, routing, the learning of routing information, and end-to-end delivery rules. The IP and IPX protocols most closely match the OSI network layer (Layer 3) and are called Layer 3 protocols because their functions most closely match OSI’s Layer 3.

• The Data Link Layer (Layer 2) is concerned with getting data across one particular link or medium. The data link protocols define delivery across an individual link. These protocols are necessarily
concerned with the type of media in use. Examples include: IEEE 802.3/802.2, HDLC, Frame Relay, PPP, FDDI, ATM, and IEEE 802.5/802.2.

• The Physical Layer (Layer 1) deals with the physical characteristics of the transmission medium. Connectors, pins, use of pins, electrical currents, encoding, and light modulation are all part of different physical layer specifications. Examples includes: EIA/TIA-232, V.35, EIA/TIA-449, V.24, RJ-45, Ethernet, 802.3, 802.5, FDDI, NRZI, NRZ, and B8ZS.

{ 0 comments }

Virtual Private Network

Virtual Private Network (VPN)

VPN is a private communication network provided over a public network, such as the Internet and is used within a company or organizations. VPN maintains fast, secure and reliable communications between computers situated at remote locations. VPN enables geographically scattered employees of organizations to communicate ‘ith each other.

Advantages

> Allows you to be at home and access your company’s computers in the same way as if you were sitting at work.

> Almost impossible for someone to tap or interfere with data in the VPN tunnel.

>If you have VPN client software on a laptop, you can connect to your company from anywhere in the world.

Disadvantages

> Setup is more complicated than less secure methods.

>The company whose network you connect to may require you to follow the company’s own policies on your home computers.

VPN goes between a computer and a network (client-to-server), or a LAN and a network using two routers (server-to-server). Each end of the connection is a VPN endpoint”, the connection betieen them is a “VPN tunnel”. When one end is a client, it means that computer is running VPN client software.

{ 0 comments }

Connectionless Network Protocol

CLNP is an ISO network layer datagram protocol. It is similar to the Internet Protocol (IP). It provides the same underlying service to the transport layer as IP in the TCP/IP environment. It is also named as ISO-IP However, the primary difference between the two is the size of the address. CLNP has an address size of 20 bytes whereas IP has an address size of 4 bytes. Thus, IP experiences an address shortage. CLNP exists in the network layer of the OSI protocol stack and thus, it provides connectionless datagram services over OSI networks. It provides the same maximum datagram size as IP and for those circumstances when the maximum packet size is greater than the size of the datagram, it provides mechanism for fragmentation.

{ 0 comments }

Connectionless Internetworking

When individual Local Area Networks (LANs) are connected together to create a Wide Area Network (WAN) Then it is called lnternetworking. These WANs may be connected to form even larger WANs. In connectionless Intemetworking, minimum two networks are connected.
Internetworking involves connecting networks that use different protocols, and thus it can be extremely complex. Internetworking is achieved using bridges, routers and gateways.
In connectionless Internetworking, data can be sent without any need to establish a connection first. Thus, there is no guarantee of delivery of data.

Connectionless Internetworking has the following advantages:
> Flexibility
> Robust
> No unnecessary overhead

The main drawback of connectionless Internetworking is that it is unreliable because of the following reasons:

> It does not guarantee delivery of data packets.

> The order of delivery is not guaranteed. Packets can take different routes to reach the destination.

{ 0 comments }

Internet Connection Sharing

The methods described earlier allow a single computer to connect to the Internet or a private network. If more than one computer wants to access the Internet simultaneously using a single Internet connection on a single system, a technique called Internet Connection Sharing (ICS) is used. Internet Connection Sharing (105) is Microsoft’s term to describe the technique of allowing more than one computer to access the Internet simultaneously using a single Internet connection on a single system. When you use ICS, you connect an entire LAN to the Internet using one computer. This connection to the Internet may be via modem, cable modem, ADSL, I5DN, leased line, or Ti. In most cases, 105 uses Network Address Translation (NAT) to aohieve this sharing.

ICS has the following components:

> DHCP Allocator — Assigns the gateway, IP addresses and name server in the local network.

> DNS Proxy — Resolves names on behalf of local network clients and forwards queries.

> Network Address Translation (NAT) — A set of private address is mapped to a set of public address. For outbound data flows, it tracks private-source lP addresses and public-destination IP
addresses. The P address information is changed and the required IP header information is edited dynamically.

> Auto-dial — Connections are dialed automatically.

> Application Programming Interfaces (APIS) — Used by programs for configuration, status and dial control.

{ 0 comments }

Routing Infrastructure

The entire structure of the routed Internetwork is called as the routing infrastructure. The different routing infrastructures are explained as follows:

1. Single-path routing infrastructure — Only a single path exists between any two networks in the Internetwork. Single-path internetworks are not fault tolerant. Dynamic router can make out if there is any fault. Till the time fault exists, the network becomes unreachable. Before the packets are delivered successfully across the downed link or router, it should be made operational.

2. Multi-path routing infrastructure — Multiple paths exist between networks in the Internetwork. When dynamic routing is used, multipath Internetworks are fault tolerant. They are more complex to configure. When distance based routing protocols are used, there is a high probability of routing loops being created during convergence.

3. Flat routing infrastructure — In a flat routing infrastructure, the routing table on every router in the network contains an entry for each network segment. When you use flat routing, the networks IDs have no network! subnet structure and cannot be summarized. RIP-based Internet Packet Exchange (IPX) internetworks use flat network addressing and have a flat routing infrastructure.

4. Hierarchical routing infrastructure — Through route summarization, groups of network IDs can be represented as a single routing table entry. They simplify routing tables and lower the amount of routing information that is exchanged, but planning is required.

{ 0 comments }

split horizon

One way to eliminate routing loops and speed up convergence is through the technique called split horizon. It is never useful to send information about a route back in the direction from which the original information came.
Another form of split horizon employs a technique called route poisoning. Route poisoning attempts to eliminate routing loops that are caused by inconsistent updates. With this technique, the router sets a table entry that keeps the network state consistent while other routers gradually converge correctly on the topology change. Used with holddown timers, route poisoning is a solution to long loops.

One way to eliminate routing loops and speed up convergence is through the technique called split horizon. It is never useful to send information about a route back in the direction from which the original information came.
Another form of split horizon employs a technique called route poisoning. Route poisoning attempts to eliminate routing loops that are caused by inconsistent updates. With this technique, the router sets a table entry that keeps the network state consistent while other routers gradually converge correctly on the topology change. Used with holddown timers, route poisoning is a solution to long loops.

{ 0 comments }

← Previous Entries

Next Entries →