The IP address you purchased would be used for external connections only. That’s the point of a Static IP address. Within that IP address, you would setup an office network and either use DHCP (which would automatically assign your networked computers a dynamic IP) or you could assign each one a specific IP within that network. You would probably end up binding that purchased IP to a server and then branching off the network from there (that way people could put files on the server and people OUTSIDE the network could also access them). Each computer you setup can be created within a specific network (when you run through the windows network setup, it asks you to specify the name of the network) and so you don’t need more than one router in order to create multiple networks. If you decide to go with this route, you can ask me for a more in depth description by emailing me at.
it’s very possible to use RFC 1918 space.
anyway 10 departments w/ 20 PCs +room to grow. The smallest subnet you can use is a /27 (32 hosts) 255.255.255.224 since the next subnet down (/28) only allows for a 16 hosts.
anyway knowing that, then you’d need two /24s, actually you’d want two /24’s that are contiguous or more specifically a /23 (512 available IP addresses) (255.255.254.0 subnet mask).
You will probably have to get IP’s from an upstream provider, it’s doubtful ARIN would give you a /23 based upon what you have described, so you would end up getting a /23 from whomever you use as your upstream provider. (Level3, AT&T, some local network service provider etc..)
anyway yes, you can break up a /23 into many /27s for the random departments..
You would anchor the /23 subnet on your router and either announce it via a routing protocol, OR have your upstream provider static route the /23 to your router. Internally you would either run multiple physical Fast/Gig Ethernet interfaces or you would have many virtual interfaces on your router to facilitate the internal interfaces. (for something this simple you’re going to be using a simple router, so in this case probably multiple virtual interfaces on your internal Ethernet Interface. ).
The concept of “Class” is no longer really valid for current day IP schemes. This sounds like some homework assignment since you’re talking Classful IP addressing so make sure you tell your teacher that I did your homework for you. Anyway speaking “classful” you can use IP space from any range, but you’re going to be using two contiguous /24 (class C) addresses to form a classless /23 etc. Make sure your teacher actually teaches you CIDR (or look at changing schools).
Comments on Need help designing a network? »
Caffeinated Content
The IP address you purchased would be used for external connections only. That’s the point of a Static IP address. Within that IP address, you would setup an office network and either use DHCP (which would automatically assign your networked computers a dynamic IP) or you could assign each one a specific IP within that network. You would probably end up binding that purchased IP to a server and then branching off the network from there (that way people could put files on the server and people OUTSIDE the network could also access them). Each computer you setup can be created within a specific network (when you run through the windows network setup, it asks you to specify the name of the network) and so you don’t need more than one router in order to create multiple networks. If you decide to go with this route, you can ask me for a more in depth description by emailing me at.
Caffeinated Content
it’s very possible to use RFC 1918 space.
anyway 10 departments w/ 20 PCs +room to grow. The smallest subnet you can use is a /27 (32 hosts) 255.255.255.224 since the next subnet down (/28) only allows for a 16 hosts.
anyway knowing that, then you’d need two /24s, actually you’d want two /24’s that are contiguous or more specifically a /23 (512 available IP addresses) (255.255.254.0 subnet mask).
You will probably have to get IP’s from an upstream provider, it’s doubtful ARIN would give you a /23 based upon what you have described, so you would end up getting a /23 from whomever you use as your upstream provider. (Level3, AT&T, some local network service provider etc..)
anyway yes, you can break up a /23 into many /27s for the random departments..
You would anchor the /23 subnet on your router and either announce it via a routing protocol, OR have your upstream provider static route the /23 to your router. Internally you would either run multiple physical Fast/Gig Ethernet interfaces or you would have many virtual interfaces on your router to facilitate the internal interfaces. (for something this simple you’re going to be using a simple router, so in this case probably multiple virtual interfaces on your internal Ethernet Interface. ).
The concept of “Class” is no longer really valid for current day IP schemes. This sounds like some homework assignment since you’re talking Classful IP addressing so make sure you tell your teacher that I did your homework for you. Anyway speaking “classful” you can use IP space from any range, but you’re going to be using two contiguous /24 (class C) addresses to form a classless /23 etc. Make sure your teacher actually teaches you CIDR (or look at changing schools).