Architecture Case Study: Subnetting, and Supernetting - Networking Workshop (part 1)

IP Address

An IP Address is a logical address assigned to a computer or network device that allows network communication via the Internet Protocol. It is the same as a person having a physical address to receive mail from. Without it, no postal service would know where to send your mail to. If a computer or network device does not have an IP address then it can't communicate with other devices on the network.

• Ipv4: An IPv4 address is a 32-bit number that is typically written in dotted decimal format. Each group serves as an 8-bit field separated by periods. Multiply 8 bit per group for 4 groups and you get 32-bits. This is important to understand as most IoT devices utilize IPv4 since it's easier to subnet and supernet with. These addresses are split into two parts, where one half contains the network bits and the other half contains the host bits. The network bits are to identify which network a computer is on. Multiple computers can have the same first half of an IPv4 address because it means they are on the same network. However, no two computers can share the same host bit information. The host bits uniquely identify a specific computer on a specific network. Meaning if two computers or network devices are on the same network with the same network bit information, they must have unique host bit information in order for any network communication to work.

Subnetting

Subnetting is the practice of breaking up an IP address into multiple smaller ones. Each IP address serves as an address to get to, so breaking them down gives your devices more specific routes to use. This is useful when one must design an IP addressing scheme for a network architecture or locating where network problems might occur. Understanding how to subnet is based on knowing which subnet mask each IP address belongs to. A subnet mask defines the range of IP addresses one can use in a given network. Subnet masks are written like: /24, /28, /30. The bigger the subnet mask number, the less host bit information that the subnet contains. Knowing this allows professionals to know whether or not they can subnet the address they are given in the first place. A /30 cannot be divided any further since only 2 host bits are allocated. Any further division would cause the network not to work. The outcome of subnetting is to get the most specific routes to one specific route or to divide up a network.

Supernetting

Supernetting is the practice of route summarization. This describes combining multiple IP addresses into one large route. This is a very important skill to learn and master because of the cloud. Most cloud service providers don't allow more than 100 routes per person or organization. That is literally nothing in the world of networking since most organizations can have over 500,000 routes. These organizations however still want to utilize the scalability and agility of the cloud regardless of how many routes are allowed. This is where route summarization comes into play. In order to summarize routes one must understand how different subnet masks describe different routes. In networking, specifically routing, there are always multiple ways to get to a route. The smaller the subnet mask, the more specific the route. In order to properly supernet we must know what we're really doing. The outcome of supernetting is to get the least specific route to a lot of specific routes or to create IP addressing schemes.

LAN Architecture

A Local Area Network, or LAN, is a collection of devices connected to each other in one physical location, such as a home, office, or building. These networks are relatively small to medium in size, and used for personal or business purposes. Any LAN architecture design must and will almost always have a subnet mask of /24. This is because as a network architect, one must calculate how many bits of an IP address should go to the network side or the host side. Allocating a /24 subnet mask provisions up to 250 different hosts, or computers to use and communicate over that network. If one allocated too many then one could be wasting thousands upon millions of usable IP addresses. A /24 subnet mask prevents the under subscribing hosts bits of IP addresses and it is best practice to use it for all LAN connections.

WAN Architecture

A Wide Area Network, or WAN, is typically a collection of a few devices connected to each other over a wide geographical area, such as a country, or region. These networks are relatively large in size, and used for college or business purposes. Any WAN architecture design must and will almost always have a subnet mask of /30. This is because as a network architect, one must calculate how many bits of an IP address should go to the network side or the host side. Allocating a /24 subnet mask provisions up to 250 different hosts, or computers to use and communicate over that network. If one allocated a /24 subnet mask for a WAN connection, that would too be wasting millions of usable IP addresses. A WAN connection only needs at least 2 host bits to work properly since typically only 2 devices are connected. A /30 subnet mask prevents oversubscribing hosts bits of IP addresses, increases security due to decreasing the surface area of attack vectors, and it is a best practice to use it for all WAN connections.

Subnetting within an Architecture

Supernetting within an Architecture

NB: This architecture is a high-level representation as the full one will be more detailed and much more complex. The intended audience is the general public.

Thank you for your time, and I hope you enjoyed this architecture study case.

Dan, the Architect.

• May the cloud be with you!