2.8 Naming


IP addresses are used by routers to forward packets to their destinations. However, IP addresses are not really human friendly. The solution to this problem is to associate a unique, human-friendly name to every host in a network. A mapping between the human-friendly name and the associated IP address is that is enabled by what is known as DNS, or Domain Name System.

The Domain Name System is essentially a distributed database of host information that is indexed by domain names. A domain name is a path in a large inverted tree that is referred to as the domain name space. DNS implements a hierarchical name space for Internet names . The logical structure of the DNS is shown in Figure 2-10.

Figure 2-10. Structure of DNS.

graphics/02fig10.gif

A domain is simply a subtree of the domain name space. At the very top of DNS is what is referred to as the "root." Domains below the root are referred to as top-level domains. The original top-level domains (TLDs) divided the Internet domain name space into seven domains, as shown in Figure 2-10. Internet Society (ISOC) has approved a few other TLDs in the last couple of years . Some of these are .cc, .nu, .tv, and .bz.

Every node in a tree has a text label (max 63 chars). Only the root has no label. The domain name of any node in the tree is the sequence of labels from that node to the root. If the root nodes label appears in the nodes domain name, it appears with a dot at the end (e.g., www.nokia.com.). An absolute domain name is written relative to the root. An absolute domain name is refered to as a fully qualified domain name (FQDN).

Domain names that are leaves of the tree normally are associated with hosts . These names may point to the IP address and mail routing information among others. Data associated with domain names are contained in resource records.

The following resource record types are the most commonly used:

  • A records implement the name-to-address mapping.

  • NS records give the domain name for a host that is running a name server.

  • CNAME gives the canonical name for a host.

  • Mail Exchange (MX) records give the domain name for a host that is running a mail server for the specified domain.

As mentioned earlier, the DNS is a distributed database. The distributed nature of this naming service is achieved via delegation. Delegation refers to assigining the responsibility for a subdomain to another organization. This means that the subdomain nokia.com, for example, is managed by Nokia and has complete control over the subdomain. The .com subdomain merely has a pointer to nokia.com.

What are the elements that make up the DNS? It is primarily comprised of two components : name servers and resolvers . Network nodes that store information about the domain name space are called name servers. These servers have complete information about some part of the domain, which is refered to as a zone. What is the difference between a zone and domain? A domain includes delegated data, whereas a zone, which is bounded, will never include delegated data. Figure 2-11 illustrates the concept of a zone and domain.

Figure 2-11. Zones and domains.

graphics/02fig11.gif

Resolvers are the clients that access name servers. Application programs on hosts that require address translation services use the resolver. The primary functions of the resolver are to

  • Query a name server.

  • Interpret responses from the name server.

  • Deliver the info to the appropriate application program that requested it.

Do resolvers have to be aware of the root servers in order to resolve names? No. Resolvers are configured with a name server that is commonly in the domain that the node is connected to. Name server configuration can happen as a result of the node acquiring configuration information through DHCP or Point-to-Point Protocol (PPP), for example. Resolvers then send their queries to this name server, which does most of the work on behalf of the resolver. Figure 2-12 shows the modus operandi of name resolution.

Figure 2-12. Name resolution operation.

graphics/02fig12.gif

In Figure 2-12, the mail program at a host is seeking the MX record of user @ee.berekeley.edu. The application program contacts the local name server that it is configured with. If the name server does not have this in its cache, it will query other name servers, and in the preceding example the querying is initiated from the root server. The local name server may be configured with the address of the root server, or it may contact another name server in its domain.

One final aspect of name server resolution is the inverse mapping of addresses to names. In the Internet's domain name space, the in-addr.arpa portion of the name space is utilized for this purpose. Because any name server only knows about a part of the overall domain, an inverse query is not guaranteed to return a valid response.



IP in Wireless Networks
IP in Wireless Networks
ISBN: 0130666483
EAN: 2147483647
Year: 2003
Pages: 164

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net