Navigation bar
  Start Previous page  10 of 35  Next page End Home  1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20  

Embedded Web Server for the CR16
National Semiconductor
Jeff Wright
ATM, FDDI, etc. are all unintelligible to an Ethernet node.  The purpose and design of the
Internet Protocol is to allow nodes sitting on these dissimilar LANs to internetwork.  It does so
by abstracting their conflicting Link layer protocols, providing a uniform communication
interface for all hosts.  This permits hosts residing on disparate networks to communicate, even
though they may speak a different L³ (Link layer language).  
This is where the IP address comes in.  Whereas every node on a LAN is uniquely identified at
the Data Link layer by its MAC address, each host on the Internet is uniquely identified by its
IP address.  The mapping of a host’s logical IP address to its physical MAC address is
maintained in what are called Address Resolution Protocol (ARP) tables.  Just like a phone book,
the ARP table provides an address (the actual physical address on a LAN) for each host’s name
(its IP, or logical address). 
Network #3
Network #2
Network #1
Figure 3. Ipv4 Hierarchical Addressing scheme
IP addresses (i.e. Ipv4) are 32-bit numbers, comprising two subfields: a network identifier and
a host identifier (also referred to as the netid and hostid).  Figure 3 illustrates this
hierarchical addressing scheme.  The netid field of the address uniquely identifies a specific
LAN, WAN, or other group of linked computers.  The hostid field of the address uniquely
identifies a host on the addressed network.  (Actually, the hostid specifies a unique NIC, or
Network Interface Card.  An individual computer usually - but not necessarily, has only one such
Version 4 of the Internet Protocol (IPv4) has been in use since 1981 and is slowly being
supplanted by IPv6.  Version 6 improves upon IPv4 in several areas, not the least of which is the White Papers Previous page Top Next page