50 Networking and Online Games: Understanding and Engineering
50 Networking and Online Games: Understanding and Engineering Multiplayer Internet Games 4.2.1.2 Classless Inter Domain Routing CIDR replaced the previous A, B and C class rules (hence, classless) [RFC1519] with a flexible value/prefix-size pair scheme for identifying networks the network number is encoded in the top bits of a 32-bit value, and the number of valid bits in the network number indicated by an integer prefix-size (Figure 4.9). Figure 4.9 shows that, in general, a prefix size of X results in a network that can theoretically contain up to 2(32-X) endpoints. A key benefit of CIDR was that variably sized networks could now be built from the old class C space. For example, 192.80.192/22 represents a single network with a 22-bit prefix and a network number of 192.80.192 equivalent to four contiguous class C networks (192.80.192.*, 192.80.193.*, 192.80.194.* and 192.80.195.*, where * represents any number between 0 and 255). In other words, it represents a single /22 network prefix in the backbone routers rather than four class C prefixes. In the absence of CIDR, the last class B address would have been assigned in early 1994. CIDR significantly slowed the growth rate of the backbone routing tables, and increased the density with which IP addresses could be packed into a 32-bit field. 4.2.1.3 Subnetting Creating a single network from multiple old class C networks is known as supernetting. The reverse, creating hierarchy within individual networks, is known as subnetting. Groups of endpoints may be aggregated into subnetworks (commonly referred to as subnets)if they are topologically localised within the scope of a larger network. Individual subnets contain endpoints whose addresses all fall under a common prefix (or subnet mask ), a prefix that is itself a subset of the class or CIDR prefix assigned to the network of which they are a part. Subnets are networks within networks that can be described by a longer (that is, more precise) prefix or mask than the one that describes the network itself. IP subnets are the lowest level of the IP routing and addressing hierarchy. Routing protocols do not concern themselves with local details within subnets. In all except the most simplistic network topologies, routers are needed in order to forward packets between subnets. Layer 2 links between routers, such as Ethernet or similar LANs, are also often referred to as subnets. However, while multiple IP subnets may run over a single link, an IP subnet cannot (by definition) span more than one link without an intervening router. Consider Figure 4.10, where Network 1 is made up of two internal subnets. The network s public identity (as advertised to the IP backbone s routers) is 128.80.0.0/16. Internally, Network 1 has two subnets each with a longer, more precise 24-bit prefix (a subnet mask of 255.255.255.0). Subnet 1 covers all addresses in the range 128.80.1.0 to 128.80.1.255, whereas subnet 2 covers addresses in the range 128.80.9.0 to 128.80.9.255. Subnet 1 and 2 may be geographically separate from each other yet owned by a common administrative entity (for example, a large company). Router R1 only advertises a Address Format (binary) Networks and Hosts nnnnnnnn.nnnnnnnn.nnnhhhhh.hhhhhhhh |<---X---->| 2X nets, 2(32-X) hosts Figure 4.9 CIDR relaxes the Network Prefix Lengths
Note: If you are looking for good and high quality web space to host and run your application check Lunarwebhost Mac Web Hosting services