Ethernet at the Data Link Layer

Ethernet at the Data Link Layer

Ethernet provides services corresponding to Layers 1 and 2 of the OSI reference model, and IEEE 802.3 specifies the physical layer (Layer 1) and the channel-access portion of the Data Link (Layer 2). In addition, IEEE 802.3 does not define a logical link control protocol but does not specify several different physical layers, whereas Ethernet defines only one.

Ethernet Addressing

Every network device has a unique physical identity that is assigned by the manufacturing vendor is called MAC address or Ethernet address. The mac address is also known as the hardware address while the IP address is the logical address of the device. The mac address is defined in the Hexa-decimal format generally. It consists of 6 byte (48 bits) where the first three bytes are used as the identity of the vendor and the last three bytes are used as the node identity. The mac address works on the mac sub-layer of the data link layer of the OSI model.

Switches give network managers the ability to increase bandwidth without adding unnecessary complexity to the network. Layer 2 data frames consist of both infrastructure content, such as end user content and MAC Media Access Controladdress also known as Ethernet address. At Data Link layer, no modification is required to the MAC address of the data frame when going between like physical layer interfaces, such as from Ethernet to Fast Ethernet. However, changes to Media Access Control (MAC) address of the data frames might occur when bridging between unlike media types such as FDDI and Ethernet or Token Ring and Ethernet.

Switches learn the mac address and built a table on the base of mac addressing of the LAN segment called Mac Address Table. The Address Resolution Protocol (ARP) is the protocol that resolves the IP addresses into MAC addresses. RARP, the Reverse Address Resolution Protocol is a reverse of ARP and resolves MAC addresses into IP addresses.

The MAC layer of the Gigabit Ethernet is similar to those of standard Ethernet and Fast Ethernet. Media Access Layer of Gigabit Ethernet should maintain full duplex and half duplex broadcasting.. The characteristics of Ethernet, such as collision detection, maximum network diameter, repeater rules, MAC addressing and so forth, will be the same of the Gigabit Ethernet. Support for half duplex Ethernet adds frame bursting and carrier extension, two functions not found in Ethernet and Fast Ethernet.

Ethernet Frames

There are various types of data frames such as Ethernet frames, Ethernet jumbo frames and snap frames etc. The following illustrates the frame fields associated with both Ethernet and Ethernet IEEE 802.3 jumbo frames.

Various fields exist for both Ethernet frames and Ethernet jumbo frames:

SDF = Start Frame Delimiter

FCS = Frame Check Sequence

The Ethernet frames fields illustrated in the table are as follows:

• Preamble: The alternating pattern of one and zeros tells receiving stations that an Ethernet or Ethernet jumbo frame is coming. The Ethernet frame includes an additional byte that is equivalent of the Start of Frame (SOF) field specified in the Ethernet IEEE 802.3 jumbo frames.
• Start of Frame (SOF): The IEEE 802.3 delimiter byte ends with two consecutive 1 bits, which serve to synchronize the frame reception portions of all stations on the LAN. SOF is explicitly specified in Ethernet. It is also known as Start Frame Delimiter (SFD).
• Source and Destination Address: The first 3 bytes of the addresses are specified by the IEEE on a vendor dependent basis. The last 3 bytes are specified by the Ethernet or IEEE 802.3 vendor. Source address should be unicast address which is the address of a single device while destination address could be unicast multicast of broadcast address of all nodes.
• Type (Etherne): After Ethernet processing is completed this upper layer receive data.
• Length (Ethernet jumbo frame): The length indicates the number of bytes of data that follows this field.
• Data (Ethernet): When processing at physical layer and link layer is completed , frame data is forward towards upper layer protocols for processing which is show in Type Field. Although Ethernet version 2 does not specify any padding (in contrast to IEEE 802.3), Ethernet expects at least 46 bytes of data.
• Data (Ethernet jumbo Frame): After physical layer and link layer processing is complete, the data is sent to an upper layer protocol, which must be defined within the data portion of the frame, if at all. When frame data is not enough and it should be minimum 64 bytes frame.
• Frame Check Sequence (FCS): It contains 4 bytes of CRC value which sending device has created and it is then calculated again by receiving device to find out damaged frames.

SNAP Frame

The SNAP frame has its own protocol field to identify the application layer protocol. This is really a way to allow an Ethernet frame to be used in an 802.3 frame. You can identify a SNAP frame because the fields of DSAP and SSAP are always AA, and the command field is always 3. To allow the proprietary protocols by application developers to be used in the LLC frame, the IEEE defined the SNAP format. It is mostly used in AppleTalk proprietary frames and not used very much. Cisco uses a SNAP frame with their proprietary protocol CDP, theCisco Discovery Protocol.

Related Articles on Ethernet Technology

• What is Ethernet?
• Understanding Ethernet
• Ethernet at the Data Link Layer
• Ethernet at Physical Layer
• Understanding Ethernet Networking
• Understanding Ethernet LAN Segmentation
• What is an Ethernet Card?
• What is an Ethernet Hub?
• What is an Ethernet Switch?
• Ethernet Cable
• Ethernet Cabling
• What Is the Ethernet Cable Pinout?
• What is a USB to Ethernet adapter?
• What is Power over Ethernet?
• What is an Ethernet crossover cable?
• Ethernet Cross Over Cable

Related Articles

1. Data Link Layer
2. MAC Address
3. Frame vs Packet
4. Ethernet Card
5. Ethernet

Cisco Routing Protocols

A router connects different networks with each other to share packets of data. To route a packet, a router needs to know:

• Destination IP addresses
• Learning sources
• Possible routes to the destination
• Best route to the destination
• Verify and maintain the routing information

Routers must learn the destinations that are not directly connected through IP routing. Routing uses a route that a network routing protocol adjusts automatically for topology or traffic changes. Routing protocols are used between routers to determine paths and maintain routing table. Once the path is determined a router can route a routed protocol such as IP while RIP, the Routing Information Protocol is an example of routing protocol.

The other examples of routing protocols are as follows:

• Destination IP Addresses
• Learning sources
• Possible routes to the destination
• Best route to the destination
• Verify and maintain the routing information
• BGP: Border Gateway Protocol
• EGP: Exterior Gateway Protocol
• EIGRP: Enhanced Interior Gateway Routing Protocol
• IGRP: Interior Gateway Routing Protocol
• IS-IS: ISO IS-IS
• OSPF: Open Shortest Path First

There are three classes of routing protocol:

• Distance Vector Routing Protocol such as RIP and IGRP
• Link State Routing Protocol such as OSPF
• Hybrid Routing Protocol such as EIGRP

Information on RIP, OSPF, EIGRP, BGP Routing Protocol

The Routing Information Protocol is an open standard based Distance-Vector routing protocol. It comes on two versions RIP v.1 and RIP v.2.

The Open Shortest Path First is an example of Link State routing protocol that supports variable length subnet masking and discontiguous subnets.

The Enhanced Interior Gateway Routing Protocol is an enhanced Distance Vector routing protocol. It is also known as hybrid routing protocols that shares the attributes of both Distance Vector and Link State. It is a scalable routing protocol, even the largest EIGRP installation in the world network is running successfully with no EIGRP offset.

EIGRP vs IGRP

The Border Gateway Protocol is an exterior gateway protocol that provides inter-autonomous system routing. It is the protocol of the internet. BGP operates in one of two modes: internal BGP and external BGP. BGP path attributes fall into 4 separate categories:

RIP: Followings are some important features of RIP:

• RIP supports Bellman and Ford algorithm.
• RIP v.1 uses hop count as a metric while RIP v2 routing protocol metric is also hop count.
• The maximum hop counts for both are 15.
• RIP v.1 features the use of broadcast updates while RIP v.2 features the use of multicast routing updates.
• RIP v.1 is a classful routing protocol that supports FLSM while RIP v.2 is a classless routing protocol that supports VLSM.
• The administrative distance of RIP is 120.RIP supports equal metric load balancing. OSPF: Followings are some important features of OSPF:
• OSPF advertise a large amount of topological information about the network includes what every metric is for every link.
• OSPF uses Shortest Path First (SPF) algorithm to calculate the metric.
• Using OSPF, routers perform CPU intensive data computations.
• OSPF discovers neighbours before exchanging routing information.
• The administrative distance of OSPF is 110.
• OSPF is a classless routing protocol that supports VLSM and CIDR.
• OSPF supports equal metric load balancing.
• The metric of OSPF is cost.EIGRP: EIGRP possesses the following features:
• It supports Variable Length Subnet Mask.
• It does not send a complete copy of its routing table to neighbours on a periodic basis.
• It can be enabled for IP as well as IPX and AppleTalk are also used as EIGRP protocols to connect EIGRP networks.
• It automatically redistributes routes with IGRP processes define in the same autonomous system.
• It performs same metric accumulation as IGRP.
• It creates topology table in addition to routing table for largest EIGRP network build.
• It converges quickly.
• Bandwidh, delay, load, reliability and MTU are used for calculating EIGRP routing metric.
• It uses dual algorithm.
• The administrative distance of EIGRP is 90.
• It is the Cisco proprietary routing protocol along with IGRP.
• It supports unequal metric load balancing by using “variance” command.The EIGRP is supposed the second generation of IGRP but it has some differences as well as some similarities. The comparison of EIGRP over IGRP is as follows:
• Both are Cisco Proprietary protocols.
• Both have same logic for equal cost paths.
• IGRP and EIGRP path selection is based on Bandwidth/Delay metric. Using some EIGRP commands maximum bandwidth can be changed as required while in IGRP it cannot be changed.
• Both have same metric accumulation, if divide the EIGRP metric by 256.
• EIGRP has fast convergence time while IGRP has a slow convergence time.
• EIGRP discovers neighbour before exchanging routing information while IGRP does not.
• IGRP sends full routing table during update while EIGRP does not, that’s why in EIGRP split horizon issue does not come.
• IGRP requires Distance Vector loop avoidance features while EIGRP does not.
• EIGRP uses Dual algorithm while IGRP uses Bellman and Ford algorithm.
• IGRP send periodic routing updates every 90 seconds while EIGRP send triggered change-based updates when there is a topology change has been occurred.BGP:
• Well-known Mandatory
• Well-known Discretionary
• Optional Transitive
• Optional non-transitiveBGP path attribute usages are as follows:
• ORIGIN (BGP 4 Type Code 1)
• AS_PATH (BGP 4 Type Code 2)
• NEXT_HOP (BGP 4 Type Code 3)
• MULTI_EXIT_DISC (BGP 4 Type Code 4)
• LOGICAL_PREFERENCE (BGP 4 Type Code 5)
• AUTOMIC_AGGREGATE (BGP 4 Type Code 6)
• AGGREGATOR (BGP 4 Type Code 7)

Related Articles

1. Routing Protocols
2. Routing Software
3. Routing
4. Routing Table
5. PPP Authentication Protocols

Understanding TCP-IP

What is TCP/IP?

TCP/IP is the most popular and important protocol suite that is highly implemented on corporate and public internetworks. The Internet communication is basically based on TCP/IP. Primarily TCP/IP is designed for government and military networks. In 90s, TCP/IP can be installed and used on Microsoft Windows operating system as required. But since late 90s, the TCP/IP is the default network protocol suite of Microsoft Windows and today, the TCP/IP is used in Microsoft NT, Microsoft 2000 and Microsoft XP and if you have any problem you have to download Microsoft TCP/IP protocol software. It is also the default networking protocol suite of Linux, Sun Solaris and other flavours of Unix along with other operating systems. TCP/IP version 6 is the latest updated version of the TCP/IP protocol suite.

TCP/IP Protocols

As you read above that TCP/IP is a protocol suite which is very popular and must required for Internet communications. A protocol suite is a bulk of protocols that works on different layers of the OSI reference model. The OSI layer model is a seven layer model that standardizes the protocols and reduces the complexity of understanding the network communication. The communication is possible through different protocols working on different layers of the model. The TCP/IP protocols work on network layer and transport layer of the OSI modelwhere TCP works on transport layer and IP works on network layer.

TCP/IP Network Layer Protocols

As we know that TCP/IP is a protocol suite which is a bulk of different 3rd and 4rth layer protocols. The 3rd layer or network layer processes the data packet by learning the IP address of source and destination. The TCP/IP network layer protocols are IP, ICMP, ARP and RARP etc.
A little description of each of these network layer protocols are as follows:

• IP: The Internet Protocol (IP) is a network layer protocol is a routed protocol that follows a route guided by any routing protocol and provides best delivery routing of packets. It carries the datagram from source to destination by learning the ip address encapsulated on the IP header of the data packet.
• ICMP: The Internet Control Message Protocol (ICMP) is a network layer protocol that provides the information about the operational status of the network. The “ping” command is used for sending and receiving the Echo request or Echo reply by using ICMP. The actual message resides inside the IP packet.
• ARP: The Address Resolution Protocol (ARP) is a network layer protocol that resolves the known IP address of the networking device into the data-link address or MAC address.
• RARP: The Reverse Address Resolution Protocol (RARP) is a network layer protocol that resolves the known MAC address of the networking device into the IP address assigned to the specific device. This is the reverse of ARP
• DHCP: The Dynamic Host Configuration Protocol (DHCP) is a network layer protocol that provides a framework for the automatic configuration of the IP addresses of the network hosts.

TCP/IP Transport Layer Protocols

At transport layer, the datagram segment is transported from source to destination by using the connection oriented or connectionless services. At this layer, logical port numbers are learned. TCP and UDP are the examples of the TCP/IP transport layer protocols.

A little description of transport layer protocols is as follows:

• TCP: The TCP (Transmission Control Protocol) is a transport layer protocol that is used for reliable but slow data transfer. It is a connection oriented protocol which means that if the packet does not arrive at the destination, it is going to be re-transmitted.

UDP: The UDP (User Datagram Protocol) is a transport layer protocol that is used for fast but unreliable data transfer. It is a connectionless protocol which means that it has no mechanism of re-transmission of the packets that does not arrive at the destination.

TCP/IP Applications

TCP/IP supports many Internet applications. Some of these applications are as follows:

• FTP: The FTP (File Transfer Protocol) is a connection oriented and reliable service that uses TCP for transferring the files between systems over the network or internetwork.
• TFTP: The TFTP (Trivial File Transfer Protocol) is an unreliable and connectionless service that uses UDP. The router uses TFTP for transferring the Cisco IOS images and the configuration files.
• HTTP: The HTTP (Hypertext Transfer Protocol) is a protocol that is used for browsing the web sites and World Wide Web communications.

Telnet and TCP/IP

The Terminal Emulation (Telnet) is a utility that provides the capability for remotely accessing the other computer. If you wants to login on a remote device such as computers, switches and routers etc, either at the same premises or from a different locations you should have to use this utility. You have to provide the ip address of remote device with the command “telnet“. The well known logical port of telnet is 23.

Linux TCP/IP Network Administration

Linux is a secure, scalable, dependable and reliable operating system based on shell programming. Linux also has a default support of TCP/IP protocol suite. The network administration of Linux is based on TCP/IP. You can use the TCP/IP tool “netconfig” for configuring the network interface card in Linux.

Alternatives to TCP/IP

Although the TCP/IP is the most popular protocol suite used in the network or internetwork communication but there are some other protocol suite also can be used as an alternative of TCP/IP. One of the alternatives is the Novell IPX/SPX protocol suite in which the IPX (Internet Packet Exchange) works at 3rd layer of OSI layer Model like IP in TCP/IP and SPX (Sequenced Packet Exchange) works at 4rth layer of OSI Model like TCP in the TCP/IP protocol suite. AppleTalk and DECNET are the other popular alternatives of the TCP/IP.

Related Articles

1. Transport Layer
2. TCP/IP Ports
3. TLS (Transport Layer Security)
4. Application Layer
5. PPP (Point-to-Point Protocol)