TCP/IP Connections Introduction User Modes Connect Via... Addressing Subnet Masks Config Router Address DNS Search Paths Hosts File 802.3 Saving Configs Troubleshooting Tools For More Info Glossary updated: 08/11/97		IP Addresses Before we talk about configuring the TCP/IP control panel for an IP address, let's talk about what an IP address is. 32-bit Address An IP address is a unique 32-bit number that identifies a particular host on an IP network. That means that each machine connected to the Internet (or internet) must have its own address; in many ways, its like how each home has its own phone number (loosely speaking). When you have 32-bits to express a number, you can have 2^32 possibilities for that number. To help understand how this works, let's say you have just a single bit to work with; one bit can be either "0" or "1", which can express "off" or "on", "false" or "true" (or "black" or "white" in the case of a monitor's color depth). If you have two bits, you can express 4 different states (or 2^2 states), because you can have 00, 01, 10, 11. If you have 8 bits, you can have 2^8 possibilities (= 256), etc. Dotted Decimal Notation So a 32-bit Internet address allows for a maximum of 2^32 possible Internet addresses, or 4,294,967,296. (The actual number is less, however, since certain addresses are reserved for other purposes.) That's a pretty big number for humans to deal with, so its usually broken down into 4 octets (8 bits apiece), and each octet is separated by a dot. The octets are also expressed as a decimal number (from 0 to 255, since we have 8 bits to play with, that's 2^8 or 256 possibilities). This is referred to as "dotted decimal" notation, and looks like "192.10.24.9". Hexidemical Notation You may also see hexidecimals used when talking about IP addresses, especially when referring to the Subnet Masks. When using Hexidecimals (sometimes prefaced with "0x" to indicate that it is a hexidecimal), each hex number, from 0 to F, indicates four bits, as shown below: Hexidecimal Binary 0 0000 1 0001 2 0010 3 0011 4 0100 5 0101 6 0110 7 0111 8 1000 9 1001 A 1010 B 1011 C 1100 D 1101 E 1110 F 1111 When you see "FF", it's equivalent to "11111111" in binary, which is equivalent to "255" in decimal notation. And FF FF FC 00 = 255.255.252.0 (third octet here would be 11111100).   IP Address = [Network Number+Node Number] Each IP address is comprised of a network number and a node number. The Class and Subnet Mask of the IP address are what define which bits indicate the network number, and which are the node. Routers use the network number portion of the IP address to route a packet to the router that serves that particular network, and then the node portion of the address is used to get the packet to the appropriate host on that network. More on how that works later...(or see ARP). When the node portion of an address is all zeroes (192.132.10.0), the reference is to the network itself, i.e., network 192.132.10. When the node portion of an address is all one's (192.132.10.255), it is a broadcast address, which means that all devices on network 192.132.10 will "listen" to this packet. Classes of IP Addresses We'll be talking about 3 Classes of Addresses used for host addresses; each class has a default subnet mask. A Subnet Mask is also a 32-bit number, a series of "1's", followed by a series of "0's". The number of 1's in the subnet mask indicate how many bits of the IP address express the network number (and the remainder would express the node number). For example, if the first 8 bits of the subnet mask were 1's (255.0.0.0), that would indicate that the first 8 bits of the IP address indicated the network number. Class A: In a Class A address, first bit begins with '0'; which means the possible range for the first octet is from 0-127 (since 128 -- 10000000--no longer begins with a "0"). Class A addresses have a default subnet mask of 255.0.0.0, which means that the first 8 bits are all 1's and the last 3 octets all zeroes. Because 1 bit is used to define the class, the remaining 7 bits can define the network address, which means there's a possibility of 127 Class A networks, and each can have approximately 2^24 nodes. Class A addresses are used for very large sites. As in the example below of IP address 17.0.0.1 with a subnet mask of 255.0.0.0: the mask tells us that the first 8 bits, or the first octet, is part of the network number, so the corresponding bits in the IP address, 00001001 (or 17 in decimal notation) indicate the network portion of the address. The bits that are "0" is the mask indicate the node bits. So the last 24 bits of the IP address--00000000.00000000.00000001 (or in decimal: 0.0.1)-- are the node portion of the address. That would make the device with this IP address node 1 on network 17. Class B: first two bits are '10'; which means the range is from 128 to 191 (since 192 is 11000000). The default subnet mask for Class B addresses is 255.255.0.0. The mask indicates a total of 16 bits (first two octets) as part of network address. But since the first 2 bits are class identifiers, that leaves 14 available for network numbers, for a possibility of up to 16,384 Class B networks, each with approximately 2^16 nodes. Class B addresses are for medium-sized organizations, such as universities. Class C: first three bits are '110'; which means the range is from 192-223 (since 224 is 11100000). The default subnet mask for Class C addresses is 255.255.255.0. Of the 24 bits used for the network address, the first 3 are used to identify the class, leaving 21 bits for the network number (2^21 = 2,097,152). But of the 2 million or so Class C networks, each can have only 2^8 (256) nodes. So Class C addresses are used for smaller organizations.
		[Intro |UserModes |ConnectVia |Addressing |Subnet Masks |Config|Routers ] DNS |Search Paths |Hosts File ]802.3 ]Saving Configs ]Troubleshooting ]Tools ]More Info ]Glossary ]