Introducing the Domain Name Service (DNS)
If you have ever connected to a website by name, you have used the Domain Name Service. The Domain Name Service is a service used on the Internet for resolving fully qualified domain names (FQDN) to their actual IP addresses. For example, let's say you were preparing to take the latest Windows 2000 certification exam. You've asked your co-workers what the best study guide available is, and they recommend you check out New Rider's web site and see what they have available. Your obvious question is "Where can I find New Rider's web site?" Now before DNS the answer would be 205.133.113.87, and if you are like most people, you'll remember that number for less than 30 seconds, and will probably never find the New Rider's site, or get that study guide you were looking for. What DNS does is put a user-friendly face on that pretty obscure numeric address. With DNS, your friend can tell you to go to www.newriders.com, and the DNS infrastructure of the Internet will translate the name to the correct address, 205.133.113.87. It's like a big phone book. You put in a name and it gives you the correct number. But life wasn't always this easy for Internet users.
Back in the early days of the Internet, when it was known as the ARPANET (Advanced Research Products Agency network) and the number of hosts on the network was less than 100, there used to be a master list of names and IP addresses called the HOSTS.TXT file. It was maintained by the Stanford Research Institute's Network Information Center (known as the SRI-NIC at the time) and it worked very well as long as the number of hosts was low, and changes were infrequent. Everyone using the network would periodically download a copy of this file and they would have a local table of names and addresses to connect to computers by name. Windows 2000 (and most TCP/IP stacks in general) still have this functionality, although it is seldom used in conjunction with the Internet any longer. This method of name resolution was great for a while, but as the number of computers grew this solution ran into a few issues, including network traffic, data consistency as well as the limitations of storing data in a flat file.
Fortunately for those of us with a limited ability to memorize strings of numbers, and limited patience with flat files that need to be updated all the time, the Internet community recognized the benefits of a dynamic name resolution system as a critical part of the infrastructure that would make up the original Internet architecture. DNS was the result, and it continues to be one of the foundations of the Internet's infrastructure.
Introducing the Domain Name Service (DNS) - In the "Real World"
OK, if you are trying to pass Microsoft Exam 70-216: Implementing and Administering a Microsoft® Windows® 2000 Network Infrastructure, you will need to understand all the finer points of implementing DNS in a Windows 2000 environment. You will need to have an understanding of Dynamic DNS, the different types of DNS records, and how to use the Microsoft Management Console to configure your DNS server. However, once you have learned all those great facts about DNS, and have successfully navigated the exam, you will have one other thing you'll need to do, if you haven't already. You'll need to implement DNS in a network environment. That's when things get more interesting.
The first thing to remember about DNS is it is used to make things easier for a user to find IP-based resources. With that thought in mind, there are generally three solutions that DNS can provide.
First, DNS can be used on the inside of your network to provide your users easier access to IP-based services like Intranet servers, application servers, proxy servers, or any other IP-based server or service. You can even map drives in many cases by DNS name, if it is supported by your operating system. A couple of things to remember. You will need at least two DNS servers to ensure a highly available DNS service. There is nothing more important to a successful DNS implementation than reliability. If you implement an unreliable DNS, your users won't use it, and you will be wasting your time. Second, make sure that the name of your internal domain and all the associated entries are consice and easy to remember. If the domain name is newriders.com, and the name of the server if thefileserveronthefirstfloornearthecopier, everyone will continue to use it's IP address. Just because you can make the name 255 characters long, doesn't mean it's a good idea.
The next place you will need a DNS server is for the outside, or Internet, side of your network. If you are going to host a web site, FTP server, or any other IP-based host on the Internet it will need an entry in an Internet DNS server. You may find that it is easier and more cost effective to let your Internet Service Provider provide DNS services, since external DNS tables generally hold a small number of entries. Make sure that your ISP uses redundant DNS servers if they are providing your DNS hosting.
Finally (and this could be the same DNS server as the first example) you will absolutely need DNS to work in conjunction with your Windows 2000 Active Directory Services implementation. There are still discussions about whether or not basing ADS on the DNS hierarchical structure was a good idea, but it's a little late for second guessing so we all need to get used to it. The reason I mention this implementation separately is that it is absolutely critical that you have a rock solid DNS architecture and implementation before deploying Windows 2000 with ADS. Without DNS, Active Directory just won't work properly.