Networking & Infrastructure

CIDR (Classless Inter-Domain Routing) notation specifies IP network blocks using a slash suffix, such as /24, indicating that the first 24 bits of the address form the network portion and the remaining bits identify hosts within that network. A /24 network has 256 total addresses, of which 254 are usable host addresses once the network and broadcast addresses are subtracted.

Subnet calculators compute four values from a CIDR block: the network address (all host bits zero), the broadcast address (all host bits one), the usable host range, and the subnet mask in dotted-decimal form. Getting any of these wrong when provisioning infrastructure causes routing failures or address overlap between segments.

Beyond single-subnet math, real network planning frequently requires aggregating multiple contiguous subnets into the smallest possible supernet (CIDR aggregation, sometimes called route summarization) to reduce the number of routes a router must hold, or splitting one larger block into several smaller subnets sized to specific host-count requirements (CIDR splitting). Both operations depend on the same bitwise arithmetic as basic subnetting, just applied in the opposite direction.

IPv6 addresses use 128 bits instead of IPv4's 32, written as eight groups of four hexadecimal digits. Because the full form is long, RFC 5952 defines a canonical compressed text representation that collapses the longest run of consecutive all-zero groups to a double colon (::) and removes leading zeros in each group. Expanding a compressed address back to its full form, or compressing a full address to its canonical short form, is a mechanical but error-prone conversion when done by hand.

A typical infrastructure planning task looks like this: you've been allocated the block 10.20.0.0/16 for a new site and need to carve it into separate subnets for servers, workstations, and IoT devices, each sized differently. Start with the CIDR Split Calculator to divide the block into right-sized ranges — for example a /24 for 250 servers, a /23 for 500 workstations, and a /26 for 60 IoT devices — and confirm none of the resulting ranges overlap.

Once the individual subnets are assigned, use the IPv4 Subnet Calculator to double-check each block's network address, broadcast address, and usable host range before entering it into router or DHCP configuration. If several allocated blocks are later found to be contiguous and no longer need to be routed separately, the CIDR Aggregator can collapse them back into a single summary route to keep the routing table small.

For IPv6 deployments, the same planning exercise uses the IPv6 Subnet Calculator to work out prefix boundaries, and the IPv6 Expansion and Compression tools to move between the full and canonical-compressed address forms when cross-referencing configuration files, DNS records, and firewall rules that may use either representation.

When a device needs to be identified at the link layer rather than the network layer — for asset inventory, DHCP reservations, or Wake-on-LAN — the MAC Address Generator produces correctly formatted addresses for lab and test environments, and the ASN Lookup and Port Number Reference tools help confirm routing ownership and standard service ports when documenting the resulting network design.

Mistake: treating the broadcast address as an assignable host address. In a /24 network like 192.168.1.0/24, 192.168.1.255 is reserved as the broadcast address and cannot be assigned to a device — a device misconfigured with this address will appear to have connectivity but will disrupt broadcast traffic for the whole segment. Fix: always calculate the usable host range explicitly (network address plus one, through broadcast address minus one) rather than assuming the full block is assignable.

Mistake: overlapping subnet allocations across sites connected by VPN or routing. Two sites independently choosing 192.168.1.0/24 will create unreachable or ambiguous routes once connected. Fix: maintain a single source of truth for allocated CIDR blocks across all connected sites before assigning a new one, and use the CIDR Split Calculator's output as documentation, not just a one-time calculation.

Troubleshooting: if a device can reach hosts inside its own subnet but not outside it, check whether its configured subnet mask matches what the network actually uses — a host configured with a /24 mask on a network actually segmented as /25 will incorrectly believe some remote hosts are local, and requests to those addresses will silently fail with no explicit error.

Troubleshooting: if an IPv6 address in a configuration file does not match what a tool or log shows, check whether one representation is compressed (using ::) and the other is fully expanded — they can be the identical address written two different ways. Normalize to one form, using RFC 5952 canonical compression as the safest default, before comparing addresses as strings.

CIDR itself is defined by RFC 4632, and was originally introduced in RFC 1519, replacing the older classful A/B/C network system with variable-length subnet masking.

IPv6 address text representation, including the canonical compression rules referenced throughout this hub, is defined by RFC 5952. The underlying IPv6 addressing architecture is defined by RFC 4291.

Private IPv4 address ranges, used in the planning example above, are reserved by RFC 1918: 10.0.0.0/8, 172.16.0.0/12, and 192.168.0.0/16.

Wake-on-LAN is not an IETF standard — it originated as AMD and HP's Magic Packet Technology specification in 1995 and was adopted as a de facto industry convention. The magic packet format (6 bytes of 0xFF followed by 16 repetitions of the target MAC address) is fixed by that original specification, not by an RFC.

If you need to route traffic across networks or the public internet, address by IP — IP addresses are hierarchical and routable, while MAC addresses are only meaningful on the local link and are not routed. If you need to uniquely identify a specific physical network interface regardless of what IP it has currently been assigned, for DHCP reservations, Wake-on-LAN, or asset tracking, address by MAC. See the MAC vs IP comparison for the full tradeoff breakdown.

If your network needs more than roughly 4 billion addressable devices, needs to eliminate NAT for peer-to-peer applications, or must comply with a mandate requiring IPv6 support, plan for IPv6 alongside IPv4 rather than IPv4 alone — most production networks today run both simultaneously (dual-stack) rather than migrating outright. See the IPv4 vs IPv6 comparison for a deeper look at when each is the better fit for a specific workload.

Wake-on-LAN (WOL) allows administrators to awaken devices remotely. It relies on a "magic packet" containing 6 bytes of 0xFF followed by 16 repetitions of the target MAC address.

Because this packet is broadcast as a raw UDP frame, routers must be configured to allow directed broadcasts for cross-subnet wake triggers. Use the Wake-on-LAN Packet Generator to build the correct packet payload, and the MAC Address Generator when you need a placeholder address for testing wake configurations before pointing them at a real device.