ARP, in the most basic sense, converts IP addresses to MAC addresses.When data arrives at a gateway, ARP is used to find the MAC address of therecipient using the given IP address. It does this by first searching theaddress in the “ARP cache.” If the IP address is mapped to a physical addresson this table, ARP will alert the gateway and the data will be sent to thecorrect machine. If the address is absent, then the protocol will send out a”request packet” throughout the network to search for the correct IP.
If themachine is found, the data will be relayed, and the cache will be updated. ARPoperates on the Network layer of the OSI model, or the Internet layer of theTCP/IP model. Cisco makes the claim that ARP classifies as both a second (DataLink) and third (Network) layer protocol because IP addressing occurs at bothlevels.
However, this classification is subject to debate. There are severalother versions of ARP as well. Inverse ARP is used by Asynchronous TransferMode (ATM) networks in which the protocol builds and ATM map and sends unicastpackets to servers at the other side of a connection. Reverse ARP,unsurprisingly, operates in the reverse of ARP.
RARP will request an IP addressfrom a server after providing a MAC address. This protocol is typically used byworkstations with no permanent storage and is overshadowed by DHCP.: IPv6 is the latestversion of the Internet Protocol (IP), designed to address the shortcomings ofIPv4. IPv4 is an older protocol used to distribute IP addresses.
Each addresswas 32 bits long, allowing for potentially 232 unique addresses. Asthe internet grew it became clear that all the addresses would eventually beexhausted, and so the Internet Engineering Task Force created IPv6 to replaceit. IPv6 utilizes 128 bit addresses, which allow for 3.4x1038addresses. IPv6 (as well as IPv4) operates at the Network layer of the OSImodel, or the Internet layer of the TCP/IP model. The Network layer containshardware like routers and switches, and establishes network communicationbetween devices.1. IPSec is the framework used to secure IPcommunications.
It does this through two services: Authentication Headerprovides authentication and checks for changes during transition, andEncapsulating Security Payload performs authentication as well as encryption.IPSec operates at the Internet layer of the TCP/IP model and provides securityfor nearly all other protocols in that model. Among others, IPSec defendsagainst data corruption and theft, credentials theft, and attacks fromuntrusted computers and networks.1. BGP is a routing protocol that exchangesrouting information for the Global Internet. Furthermore, it is the protocolthat provides connections between Internet Service Providers (ISP). BGP isutilized on a much larger scale than is typically relevant for an end user.
Theprotocol routes based on paths and policies and boasts incredible scalabilitywhen compared to other routing protocols. For this reason, ISPs use it toconnect to one another, and some very large corporations use it when they needto connect through multiple ISPs. BGP can be found on the Application layer ofthe TCP/IP suite, or the Transport layer of the OSI model. The BGP as describedis sometimes referred to as the Exterior (or External) Border Gateway Protocol(EBGP). When BGP exchanges routes within an Autonomous Network, then it iscalled Interior Border Gate Protocol (IBGP)UDP is an unreliable connectionless protocol that sends and receivespackets with very little functions for error checking or recovery.
UDP operatesat the Transport layer of both the TCP/IP and OSI models. UDP is typicallydefined as a contrast to TCP. TCP provides error recovery mechanisms for theuser to ensure integrity, at the cost of increased date overhead and timedelay. UDP, on the other hand, does no perform error checking and does notrequest acknowledgement from the recipient. The result is that packets mayappear out of order of go missing entirely. The benefit of this method is thatit is far faster and more efficient, and thus is well suited for real-timecommunications like VoIP, video streaming, or online gaming.
1. POP3 is an internet standard used to remotelyretrieve email through TCP/IP. POP3 allows a local client to collect theiremail from a server and store it locally.
However, it is exclusively used forretrieving email. The protocol for transferring mail between servers is theSimple Mail Transfer Protocol). POP3 is considered to be mostly obsolete now,having been replaced by the Internet Message Access Protocol (IMAP).
IMAP doesnot download all of a client’s messages automatically, and instead allows themto read the message directly from the service. This is usually faster and moreefficient, with the added benefit of allowing synchronization across multipledevices. POP3 and IMAP both operate on the Application Layer of the TCP/IPsuite.1.
HTTP is an adaptation of HTTP, with theaddition of a Secure Socket Layer (SSL) for increased security. HTTP is thecommunication protocol that connects web servers over the internet. When a usertries to connect to a website through a browser, they submit an HTTP request tothe server. In turn, the server sends a response message and provides resourcesto the client.
However, this method is very vulnerable to attacks like theman-in-the-middle attack, so HTTPS has become the new standard. Using SSLencryption, the data sent back and forth between the client and server becomeillegible to anyone eavesdropping. HTTP and HTTPS are both on the Applicationlayer of the TCP/IP model, or the Sessions layer of the OSI model.1.
DHCP is a protocol used to dynamically assignIP addresses, as well as other configurations, to devices on a network. DHCPperforms this task automatically rather than needing to be managed by networkadmins. Furthermore, DHCP scales very well, allowing it to be implemented onsmall and large networks alike. Beyond IP addresses, DHCP will also configure adevice’s default gateway, DNS server info, and subnet mask. When a device firstconnects to a network, it broadcasts a DHCP query requesting this information.
In return, a DHCP server will respond to perform the necessary configurations. Thisprotocol operates on the Application layer of TCP/IP, or the Sessions layer ofOSI.SLIP is an outdated Internet link protocol for connection through serialports or modems. It provides neither address nor error control, and as a resulthas been rendered almost completely obsolete. When transmitting data, SLIPsends the data frame followed by a single END transmission character. SLIP ispreferred on microcontrollers for this reason, as there is very littleoverhead. For practically all other applications, however, SLIP has beenreplaced by the Point-to-Point Protocol (PPP).
PPP is preferred for datatransmission over the internet, as it features datagram encapsulation, aprotocol for communication configuration and testing, and integration controlfor the protocols of upper layers. SLIP and PPP can both be found on the DataLink layer of the OSI model, or the Network Interface Layer of the TCP/IPmodel.1. : ICMP is a support protocol for the TCP/IPsuite. ICMP’s main function is to report errors to the source IP address whenthere is an issue with packet delivery. When such an error occurs, the protocolwill generate a message to send back, usually explaining what went wrong. ICMPis also capable of giving routing suggestions in the case of unresponsive hops.
An ICMP header is composed of: The major type identifying the message, theminor code revealing more info about the type, and the checksum for errordetection. ICMPv6 is a more recent version of the protocol designed as anintegral function of IPv6. It features a number of new options and features,like the Neighbor Discover Protocol (NDP) and the Multicast Listener DiscoveryProtocol (MLD). ICMP operates on the network layer of the OSI model and theInternet layer of the TCP/IP model.