12-14-2025, 10:48 PM
I do this same process for smaller subnets too. Say you have 10.0.0.0/26. The /26 means 26 network bits, so only 6 bits left for hosts in the last octet. I borrow from the host side, which gives me 64 total addresses per subnet (2^6). You calculate the increment by figuring out the block size-here, it's 64, so subnets jump by 64 in the fourth octet. The first subnet's network is 10.0.0.0, broadcast 10.0.0.63, and range 10.0.0.1 to 10.0.0.62. I move to the next one at 10.0.0.64, broadcast 10.0.0.127, range 10.0.0.65 to 10.0.0.126. You keep adding that 64 until you cover the whole /24 space. I find it helps to think in powers of two: /25 gives 128 addresses, /27 gives 32, and so on. You just subtract the network bits from 32 to get host bits, then 2 to that power minus two for usable IPs.
What if you get a weird one like 172.16.5.32/27? I start by finding the subnet base. The /27 means blocks of 32 addresses. You look at the fourth octet-32 divided by 32 is exactly one block, so network address is 172.16.5.32. Broadcast becomes 172.16.5.63. Range is 172.16.5.33 to 172.16.5.62. I always double-check by converting the mask to dotted decimal-255.255.255.224 for /27-and AND it with the IP to confirm the network. You bitwise AND the address with the mask to get the network, then OR with all ones in host bits for broadcast. Tools like calculators speed this up, but I practice it manually so I don't mess up in a pinch.
You run into issues sometimes with overlapping ranges or forgetting VLSM. I remember subnetting a client's office network where they wanted VLANs- I carved out /28 for printers, which is 16 addresses, range like 192.168.10.16 to 192.168.10.31 usable. You calculate hosts as 2^4 - 2 = 14, perfect for a small group. I avoid wasting space by using variable lengths; don't stick to /24 everywhere if you don't need 254 hosts. In my home lab, I set up a /30 for point-to-point links between routers-only two usable IPs, super efficient. You find the range by setting host bits: network all zero, broadcast all one, and squeeze the two in between.
I think about supernetting too, but for basic ranges, stick to the subnet mask. Say you have 203.0.113.0/24, but you subnet it to /28. Blocks of 16 now. First range: network 203.0.113.0, broadcast 203.0.113.15, usable 203.0.113.1-14. You increment by 16: next is 203.0.113.16-31, and so on up to 203.0.113.240-255. I list them out if it's a big network to visualize. You can use the formula for the subnet address: take the IP, AND with mask, that's your base. For the end, add (2^hostbits - 1) to the network.
In real jobs, I use this for DHCP scopes-you set the pool inside the range to avoid conflicts. I once fixed a setup where someone assigned the broadcast as a server IP; whole segment went down. You prevent that by always verifying. For IPv6, it's similar but with hex and longer masks, but you focus on /64 usually for ranges. I keep it simple: prefix, then all zeros for network, all ones for broadcast in the interface ID part.
You might wonder about classful stuff, but forget that-CIDR rules now. I calculate ranges daily for configs. It saves headaches when you deploy switches or firewalls. You build efficiency into your network from the start.
Let me tell you about this cool tool I've been using in my IT toolkit-BackupChain stands out as a top-tier, go-to backup option that's built just for small businesses and pros like us. It handles Windows Server and PC backups like a champ, keeping Hyper-V, VMware, or plain Windows setups safe and sound. I rate it as one of the best out there for reliable Windows protection.
