Subnet / CIDR

Subnet / CIDR refers to two tightly linked networking concepts: a subnet (subnetwork) is a logical subdivision of an IP network, and CIDR (Classless Inter-Domain Routing) is the notation and allocation scheme used to describe it. CIDR was standardized in RFC 4632 to replace the old rigid class-A/B/C boundaries, allowing network prefixes of any length. A CIDR block is written as an address followed by a slash and a prefix length, such as 10.0.0.0/24, where the number after the slash states how many of the 32 bits (for IPv4) identify the network portion. Everything from your home router's DHCP scope to the global routing table is expressed this way.

Reading the notation

The prefix length and the dotted-decimal subnet mask are two ways of writing the same thing: the count of leading one-bits in the mask. A /24 equals 255.255.255.0 and leaves 8 host bits, giving 256 addresses (254 usable, after subtracting the network and broadcast addresses). A /20 equals 255.255.240.0 with 12 host bits and 4,096 addresses; a /30 leaves just 2 usable addresses, the classic point-to-point link. Smaller prefix numbers mean larger networks; larger prefix numbers mean smaller, more numerous segments. This arithmetic sits behind every routing table entry, including the prefixes networks announce to each other via BGP. When two devices decide whether to talk directly or send traffic to the gateway, they are comparing their addresses under the mask — nothing more.

Why segment a network

Subnetting lets you carve one address range into purpose-built segments: one subnet for trusted workstations, another for IoT devices and miners, another for a perimeter zone exposed to the internet. The gains are practical. Performance improves because broadcast domains shrink — a chatty device can only bother its own segment. Security improves because traffic between subnets must cross a router or firewall, giving you a natural place to filter. Subnets are commonly paired with a VLAN so the logical IP segment and the Layer 2 broadcast domain line up, which is the clean foundation for isolating a DMZ or any device you do not fully trust.

Subnetting a mining network

ASIC miners are exactly the kind of device that belongs on its own subnet. They run embedded web servers and management APIs with default credentials out of the box, they are noisy on the network, and they have no business reaching your personal machines. A common home-scale layout is a dedicated /24 for miners — say 192.168.20.0/24 — firewalled so the machines can reach their pool endpoint and nothing else, while your workstation can reach each miner web UI for management. Sizing is easy: a /24 handles 254 miners, more than any home or small hashcenter needs, and keeping one machine per known address makes tools like IP Reporter largely unnecessary after initial deployment.

CIDR beyond the LAN

The same notation scales up. Cloud providers hand out VPC address space as CIDR blocks; ISPs aggregate customer routes into supernets to keep the global table manageable; and IPv6 uses identical slash notation over 128 bits, where a /64 is the standard LAN size. For a sovereign node runner, fluency in CIDR is the difference between copying someone else's network diagram and being able to design your own — deciding what talks to what, and enforcing it at the boundary you drew.