According to the scenario, the company will have a single, physical location with approximately145 hosts on the network. In this case, I would suggest using a class C address such as 192.168.0.0. Reason why is because you can put all the devices on a single subnet with the subnet mask 255.255.255.0/24. This would allow for 253 usable host addresses, which is more than needed in this case.
Whether you should dynamically or statically assign those IP’s goes, I suggest assigning it dynamically. Unless you don’t mind MANUALLY managing 218 hosts, you should definitely go with DHCP. The router will be well capable of handing out addresses for 145 hosts.
Since there is only one location used in this case, only one network should be used as it will only complicate things to have more than one network.
Since DHCP is being used as the suggestion, whether we should use a router and/or firewall, or utilizing Windows Server…the answer is yes. The router for dishing out the IP addresses and the firewall should be used for security purposes. Using Windows Server just for DHCP is rather overkill.
Using these suggested methods should also accommodate for at least 50% growth within the next few years.
Windows 2008 Network Services
To understand the functions of DNS and DHCP, first we must acknowledge their necessities. Humans in general are fairly decent at remembering certain numbers, like home addresses, phone numbers, and Social Securities. And even those simple tasks are becoming difficult for us because of technology doing all the remembering for us. We are becoming lazier and more dependent on computers. If we can’t even do these menial tasks efficiently, how would you expect us to remember dozens of IP addresses for websites? This brings to a solution. Also known as name resolution, it allows us to enter the name of a website, and then the resolution service finds the IP address for that website and directs the request to