IPv4 Block Sizes Explained: When a /22 Beats a /24

Block size matters more than most people expect when buying IPv4. It decides how many usable addresses you actually get, how cleanly you can announce them, and how painful your next expansion will be. I’ve watched this play out in data center rollouts and multi-site enterprise networks alike: someone buys a /24 because it’s the safe, familiar choice, and eighteen months later they’re back in the market scrambling for more space. The difference between a /24 and a /22 isn’t just capacity, either. It has real financial and operational consequences that tend to compound over time. Here’s how the two sizes actually compare, and why the small price premium on a /22 usually earns itself back faster than buyers anticipate. If you’re shopping right now, IP4 Market lists verified blocks in all standard sizes with transparent pricing.

IPv4 Block Sizes: A Quick CIDR Refresher

CIDR, or Classless Inter-Domain Routing, defines block sizes by prefix length. Shorter prefix, bigger block. Simple as that:

• /24: 256 total addresses, 254 usable once you set aside the network and broadcast addresses.
• /23: 512 total, 510 usable.
• /22: 1,024 total, 1,022 usable.
• /21: 2,048 total, 2,046 usable.

Prefix length shapes routing too, and this is the part people underestimate. A /24 is the longest prefix guaranteed to propagate across the entire global routing table, which explains why it became the default for so many buyers. Go longer than a /24 and you risk being filtered by operators with strict prefix-length limits. Go shorter and your space aggregates more nicely, but you’re writing a bigger check upfront. Alignment matters as well: a /22 has to sit on a natural /22 boundary. Verify that before you buy anything. The good marketplaces, IP4 Market included, check block legibility and RIR eligibility before a transaction closes, which protects you from misaligned or contested space.

The Case for the /24: Simplicity and Granular Control

The /24 earned its status honestly. It’s the smallest block that routes everywhere, and that makes it a good fit for discrete, self-contained uses:

• Dedicated infrastructure: one /24 covers a point-of-presence, a DMZ segment, or a customer-facing range without leftovers to manage.
• Geographic separation: assign a /24 per data center or region and your geofeed and reverse DNS work stays simple.
• Budget-sensitive acquisitions: with /24s historically trading around $25–$35 per address, you’re looking at roughly $6,000–$9,000 of capital. A /22 costs about double.

Granularity is where the /24 really shines, in my experience. Smaller blocks are easier to sell, transfer, or sublease one at a time, and they give network teams tidy per-site boundaries for ACLs, BGP communities, and DDoS mitigation zones. If your growth has a ceiling, or your requirements are nailed down and unlikely to move, a /24 is a perfectly rational pick. No shame in it.

Why a /22 Often Beats a /24 for Growing Networks

But growth changes the math. Start with raw capacity: 1,022 usable addresses versus 254 means you can absorb four sites, four VLAN groups, or a fourfold subscriber jump without ever going back to the transfer market. That last part matters more than it sounds. Networks stitched together from multiple /24s tend to accumulate operational debt, extra BGP announcements, messier route aggregation, longer ACL sets, and every one of those is another place for a misconfiguration to hide.

Pricing tilts in the /22’s favor too. Per-address costs for /22 blocks typically sit at or slightly below the per-address cost of /24s, mostly because ISPs and cloud operators want larger contiguous blocks and bid them up. In practice buyers often end up with more addresses for proportionally less money. And there’s a one-way door here worth dwelling on: you can always announce a /22 as four /24s if you need to. You cannot merge four separately acquired /24s into a routable /22 unless they happen to be contiguous and aligned, which they almost never are.

Routing, BGP Announcements, and Cost Implications

Every prefix you announce consumes resources on routers somewhere in the global routing table. A /22 and a /24 each occupy a single BGP entry when announced as one aggregate. The real difference shows up at the edges of your network. With a /22 you can deaggregate into four /24s for traffic engineering — steering traffic across multiple upstreams, balancing load between data centers, or isolating a noisy customer — while still advertising the aggregate to keep the global table tidy.

A lone /24 gives you none of that. Need to split it across two sites? You’re forced to announce two /24s or carve it up with more specific routes, which means more route churn and a higher risk of route leaks. RPKI matters here too. Both block sizes validate identically, but maintaining ROAs for a single /22 is administratively simpler than juggling several /24s picked up at different times from different holders.

On cost, routing fees are largely agnostic to prefix length — most transit providers and IXPs charge per port, per ASN, or per announced prefix. That said, some blacklist and DDoS mitigation services price per announced netblock, and CDN partners sometimes demand a minimum aggregate size for dedicated capacity. A /22 often unlocks tiered pricing a single /24 simply can’t reach, which effectively lowers your per-address cost across the contract.

IP Reputation and Spam Filtering: Does Block Size Matter?

Reputation systems like Spamhaus, Barracuda, and Microsoft SNDS usually score IP addresses individually. But block-level listings do happen. When a /22 gets listed at the block level, the damage spreads across 1,024 addresses; a /24 listing hits only 256. In practice, though, the bigger block is often safer. Here’s why:

• Larger allocations historically belong to established organizations with cleaner usage patterns, and reputation engines weigh that favorably.
• A /24 dedicated to a single mail server carries all of that server’s sending behavior. A /22 lets you separate mail, web, and application traffic into distinct sub-blocks, containing any reputation damage before it spreads.
• Some receivers apply stricter scrutiny to small, recently transferred blocks — statistically, those are associated with spam operations.

Whatever size you go with, check the block against the major DNSBLs and BGP hijack history databases before you buy. Legacy space with a troubled past can take months to rehabilitate. Prefix length won’t save you there.

Choosing the Right Block Size: Final Recommendations

1. Under 200 addresses needed: a /24 is usually enough, and cheaper on the open market.
2. 200–700 addresses with growth plans: a /22 gives you headroom, routing flexibility, and better long-term economics.
3. Multiple services or tenants: go with the /22 so you can partition cleanly without carving up a scarce resource.
4. Pure investment holding: /22s and larger blocks tend to be more liquid and appreciable than single /24s.