Subnet Calculator: IPv4 & IPv6 CIDR, VLSM and route aggregation

Paste one network or a hundred. Every line returns its network and broadcast address, usable host range, subnet and wildcard mask, host count and address type, recalculated as you type, then exports to CSV or JSON.

  • Bulk modeone network per line, IPv4 and IPv6 mixed freely, every line calculated at once, not one subnet at a time.
  • Exact IPv6 maths128-bit integer arithmetic, so a /48 or /64 reports its true address count instead of an approximation.
  • VLSM splitterdivides a parent block into right-sized child subnets from a list of host counts, packed largest-first.
  • CIDR aggregationcollapses a route list into the fewest supernets covering exactly the same addresses.
  • CSV, JSON and share linksexport the whole result table, or send a link that reloads the exact same list.
  • Nothing leaves your browserall maths runs client-side. No upload, no account, and it keeps working offline.

Subnet calculator

Enter one or more networks in CIDR (10.0.0.0/24), IPv6, or IP + mask form. For each one you get the network and broadcast address, the usable host range, the mask, and how many hosts it holds, recalculated the instant you type.

Results
3 networks · updates live
SubnetNetworkSubnet maskWildcard maskUsable rangeUsable hostsType
255.255.255.00.0.0.255192.168.1.1 - 192.168.1.254254Private
255.0.0.00.255.255.25510.0.0.1 - 10.255.255.25416,777,214Private
ffff:ffff:ffff::::ffff:ffff:ffff:ffff:ffff2001:db8::1 - 2001:db8:0:ffff:ffff:ffff:ffff:fffe≈1.20 × 10²⁴Documentation
192.168.1fixed by the prefix64split inside this octet0host space/26prefix length
3 networks2 IPv4, 16,777,472 addresses1 IPv6, ≈1.20 × 10²⁴ addresses0 errors

IP address pool reference

The reserved IPv4 and IPv6 ranges, with what each one is for and whether it routes on the public internet. Select any block to load it into the calculator above. Values are taken from the IANA special-purpose address registries.

Private / internal

blockaddressespurposereference
16,777,216Private-UseInternal networks. The largest RFC 1918 block, usual choice for data centres and large campuses.RFC 1918not routed
1,048,576Private-UseInternal networks. Covers 172.16.0.0 to 172.31.255.255, which is the range most often mis-stated as 172.16 to 172.16.RFC 1918not routed
65,536Private-UseInternal networks. The default on home and small office routers.RFC 1918not routed
≈2.65 × 10³⁶Unique-LocalIPv6 equivalent of RFC 1918. In practice only fd00::/8 is used, with a randomly generated 40-bit global ID; fc00::/8 is unassigned.RFC 4193not routed

Shared address space

blockaddressespurposereference
4,194,304Shared Address SpaceCarrier-grade NAT between subscriber and provider. Not private space: do not use it as extra RFC 1918 room, because your ISP may already be using it.RFC 6598not routed

Loopback

blockaddressespurposereference
16,777,216LoopbackThe host itself. The whole /8 loops back, not just 127.0.0.1.RFC 1122not routed
1Loopback AddressThe host itself. IPv6 spends a single address here rather than a whole block.RFC 4291not routed

Link-local

blockaddressespurposereference
65,536Link LocalSelf-assigned when DHCP fails (APIPA). An address from here usually means the DHCP server did not answer.RFC 3927not routed
≈3.32 × 10³⁵Link-Local UnicastAutomatically configured on every IPv6 interface and required for neighbour discovery. Always present, never routed off the link.RFC 4291not routed

Multicast

blockaddressespurposereference
268,435,456MulticastIPv4 multicast, the old class D. 224.0.0.0/24 is link-local control traffic such as OSPF and VRRP.RFC 5771not routed
≈1.32 × 10³⁶MulticastIPv6 multicast. The next 4 bits are flags and the 4 after that are scope, so ff02:: is link-local and ff0e:: is global.RFC 4291not routed

Documentation and examples

blockaddressespurposereference
256Documentation (TEST-NET-1)Examples and documentation. Safe to print, never routed.RFC 5737not routed
256Documentation (TEST-NET-2)Second documentation block, for examples needing two distinct networks.RFC 5737not routed
256Documentation (TEST-NET-3)Third documentation block.RFC 5737not routed
≈7.92 × 10²⁸DocumentationThe IPv6 documentation prefix, seen in nearly every IPv6 example ever written.RFC 3849not routed
≈3.24 × 10³²DocumentationAdded in 2024 because a single /32 was too small for realistic multi-site examples.RFC 9637not routed

Globally routable

blockaddressespurposereference
≈4.25 × 10³⁷Global UnicastThe whole publicly routable IPv6 range. Every address an RIR assigns comes from here, which is one eighth of the address space.RFC 4291routes globally

IPv4 / IPv6 transition

blockaddressespurposereference
4,294,967,296IPv4-mapped AddressCarries an IPv4 address inside an IPv6 socket, so ::ffff:192.0.2.1 is 192.0.2.1. Appears in logs on dual-stack hosts.RFC 4291not routed
4,294,967,296IPv4-IPv6 TranslationThe well-known prefix for NAT64, letting IPv6-only clients reach IPv4 servers.RFC 6052routes globally
≈5.19 × 10³³6to4Legacy automatic tunnelling. Deprecated in practice; reachability depends on a relay.RFC 3056depends on relay
≈7.92 × 10²⁸TeredoTunnels IPv6 over UDP through NAT. Reachability depends on the relay.RFC 4380depends on relay

Benchmarking

blockaddressespurposereference
131,072BenchmarkingReserved for testing network devices. Traffic must not leave the test rig.RFC 2544not routed
≈1.20 × 10²⁴BenchmarkingIPv6 equivalent for device testing.RFC 5180not routed

Other reserved

blockaddressespurposereference
16,777,216This networkSource address only, and only before the host knows its own address. 0.0.0.0/32 alone means this host.RFC 791not routed
268,435,456ReservedThe old class E. Reserved indefinitely and dropped by most stacks, so it is not the spare space it looks like.RFC 1112not routed
1Limited BroadcastBroadcast to the local link only. Never forwarded by a router.RFC 8190not routed
256IETF Protocol AssignmentsReserved for protocol machinery, including the 192.0.0.0/29 DS-Lite range.RFC 6890not routed
1Unspecified AddressNo address yet. Seen as the source during duplicate address detection.RFC 4291not routed
≈5.19 × 10³³Segment Routing (SRv6) SIDsSegment identifiers for SRv6, allocated in 2024. Forwardable inside an operator's domain but not globally reachable.RFC 9602not routed
≈1.84 × 10¹⁹Discard-Only Address BlockA managed black hole: routes here are meant to be discarded, which is used for remotely triggered filtering.RFC 6666not routed

TCP and UDP port reference

133 ports engineers actually look up, with the IANA service name, what the port is used for in practice, and a warning where the protocol sends credentials in the clear. Search by number, by service, or by what a firewall log called it.

133 of 133 ports

Web and proxies8

portprotoserviceused for
80tcp/udp/sctphttpWorld Wide Web HTTPHTTPUnencrypted. Use 443 (HTTPS) in production.
443tcp/udp/sctphttpshttp protocol over TLS/SSLHTTPS
631tcp/udpippIPP (Internet Printing Protocol)IPP printing
1080tcp/udpsocksSocksSOCKS proxy
3128tcp/udpndl-aasActive API Server PortSquid proxynot the IANA assignment
8080tcp/udphttp-altHTTP Alternate (see port 80)HTTP alternateCommon for proxies, app servers and containers behind a reverse proxy.
8443tcp/udppcsync-httpsPCsync HTTPSHTTPS alternatenot the IANA assignmentWidely used as a second HTTPS port; IANA assigns it to PCsync.
9100tcp/udppdl-datastreamPrinter PDL Data StreamPrinter raw / JetDirect

Mail7

portprotoserviceused for
25tcp/udpsmtpSimple Mail TransferSMTP relayUnencrypted. Use 587 with STARTTLS, or 465 implicit TLS.
110tcp/udppop3Post Office Protocol - Version 3POP3Unencrypted. Use 995 (POP3S).
143tcpimapInternet Message Access ProtocolIMAPUnencrypted. Use 993 (IMAPS).
465tcpsubmissionsMessage Submission over TLS protocolSMTP submission over TLS
587tcp/udpsubmissionMessage SubmissionSMTP submission
993tcpimapsIMAP over TLS protocolIMAP over TLS
995tcp/udppop3sPOP3 over TLS protocolPOP3 over TLS

File transfer and sharing13

portprotoserviceused for
20tcp/udp/sctpftp-dataFile Transfer [Default Data]FTP dataUnencrypted. Use SFTP over 22, or FTPS on 989/990.
21tcp/udp/sctpftpFile Transfer Protocol [Control]FTP controlUnencrypted. Use SFTP over 22, or FTPS on 989/990.
69tcp/udptftpTrivial File TransferTFTPUnencrypted. No authentication at all. Keep it off routable networks.
139tcp/udpnetbios-ssnNETBIOS Session ServiceNetBIOS sessionLegacy SMB transport.
445tcp/udpmicrosoft-dsMicrosoft-DSSMB / CIFSWindows file sharing. Never expose to the internet.
548tcp/udpafpovertcpAFP over TCPAFP
873tcp/udprsyncrsyncrsync
989tcp/udpftps-dataftp protocol, data, over TLS/SSLFTPS data
990tcp/udpftpsftp protocol, control, over TLS/SSLFTPS control
2049tcp/udp/sctpnfsNetwork File System - Sun MicrosystemsNFS
3260tcp/udpiscsi-targetiSCSI portiSCSI target
3690tcp/udpsvnSubversionSubversion
9418tcp/udpgitgit pack transfer serviceGit protocolUnencrypted. Unauthenticated and unencrypted; prefer SSH or HTTPS.

Remote access and management9

portprotoserviceused for
22tcp/udp/sctpsshThe Secure Shell (SSH) ProtocolSSH / SFTP / SCP
23tcp/udptelnetTelnetTelnetUnencrypted. Credentials sent in the clear. Use 22 (SSH).
623udpasf-rmcpASF Remote Management and Control ProtocolIPMI / BMCOut-of-band server management. Keep it on an out-of-band management network.
2000tcp/udpcisco-sccpCisco SCCPCisco SCCP
3389tcp/udpms-wbt-serverMS WBT ServerRDPMicrosoft Remote Desktop. A top ransomware entry point when exposed.
5555tcp/udppersonal-agentPersonal AgentAndroid ADBnot the IANA assignmentIANA assigns this to personal-agent.
5900tcp/udprfbRemote FramebufferVNCUnencrypted. VNC auth is weak; tunnel it over SSH.
5985tcp/udpwsmanWBEM WS-Management HTTPWinRM HTTPUnencrypted. Use 5986 for WinRM over TLS.
5986tcp/udpwsmansWBEM WS-Management HTTP over TLS/SSLWinRM HTTPS

Directory and authentication11

portprotoserviceused for
49tcp/udptacacsLogin Host Protocol (TACACS)TACACS+
88tcp/udpkerberosKerberosKerberos
113tcp/udpauthAuthentication Serviceident
389tcp/udpldapLightweight Directory Access ProtocolLDAPUnencrypted. Use 636 (LDAPS) or StartTLS.
464tcp/udpkpasswdkpasswdKerberos password change
636tcp/udpldapsldap protocol over TLS/SSL (was sldap)LDAPS
1645tcp/udpsightlineSightLineRADIUS (legacy)not the IANA assignmentPre-standard RADIUS port, still seen on older kit. IANA assigns it to SightLine.
1812tcp/udpradiusRADIUSRADIUS authentication
1813tcp/udpradius-acctRADIUS AccountingRADIUS accounting
3268tcp/udpmsft-gcMicrosoft Global CatalogGlobal Catalog
3269tcp/udpmsft-gc-sslMicrosoft Global Catalog with LDAP/SSLGlobal Catalog over TLS

Databases and caches10

portprotoserviceused for
1433tcp/udpms-sql-sMicrosoft-SQL-ServerMicrosoft SQL Server
1434tcp/udpms-sql-mMicrosoft-SQL-MonitorSQL Server Browser
1521tcp/udpncube-lmnCube License ManagerOracle DB listenernot the IANA assignmentUniversally used by Oracle; IANA assigns it to nCube License Manager.
2379tcpetcd-clientetcd client communicationetcd client
3306tcp/udpmysqlMySQLMySQL / MariaDB
5432tcp/udppostgresqlPostgreSQL DatabasePostgreSQL
6379tcpredisAn advanced key-value cache and storeRedisNo auth by default in older builds. Never expose it.
9200tcp/udpwap-wspWAP connectionless session serviceElasticsearch HTTPnot the IANA assignmentIANA assigns it to WAP session service.
11211tcp/udpmemcacheMemory cache serviceMemcachedUDP mode has been abused for reflection attacks; disable it.
27017tcpmongodbMongo database systemMongoDB

Core network services21

portprotoserviceused for
53tcp/udpdomainDomain Name ServerDNSUDP for queries, TCP for zone transfers and large responses.
67tcp/udpbootpsBootstrap Protocol ServerDHCP server
68tcp/udpbootpcBootstrap Protocol ClientDHCP client
111tcp/udpsunrpcSUN Remote Procedure CallRPC portmapperFronts NFS and other RPC services.
123tcp/udpntpNetwork Time ProtocolNTP
135tcp/udpepmapDCE endpoint resolutionMS RPC endpoint mapper
137tcp/udpnetbios-nsNETBIOS Name ServiceNetBIOS name
138tcp/udpnetbios-dgmNETBIOS Datagram ServiceNetBIOS datagram
161tcp/udpsnmpSNMPSNMPUnencrypted. v1 and v2c community strings are plaintext. Use SNMPv3.
162tcp/udpsnmptrapSNMPTRAPSNMP trap
427tcp/udpsvrlocServer LocationSLP
514udpsyslogSyslogUDP 514 is syslog; TCP 514 is the old rsh shell service.
546tcp/udpdhcpv6-clientDHCPv6 ClientDHCPv6 client
547tcp/udpdhcpv6-serverDHCPv6 ServerDHCPv6 server
853tcp/udpdomain-sDNS query-response protocol run over TLSDNS over TLS
1900tcp/udpssdpSSDPSSDP / UPnPReflection-attack amplifier. Block inbound at the edge.
3702tcp/udpws-discoveryWeb Service DiscoveryWS-Discovery
5353tcp/udpmdnsMulticast DNSmDNS / Bonjour
5355tcp/udpllmnrLLMNRLLMNRFrequently abused for credential relay; disable on Windows networks.
6514tcp/udpsyslog-tlsSyslog over TLSSyslog over TLS
7547tcp/udpcwmpBroadband Forum CWMPTR-069 CWMPISP remote management of CPE.

Routing and network control14

portprotoserviceused for
179tcp/udp/sctpbgpBorder Gateway ProtocolBGP
520udprouterlocal routing process (on site); uses variant of Xerox NS routing information protocol - RIPRIP
521tcp/udpripngripngRIPng
646tcp/udpldpLDPMPLS LDP
830tcp/udpnetconf-sshNETCONF over SSHNETCONF over SSH
1985tcp/udphsrpHot Standby Router ProtocolHSRP
2055tcp/udpiopIliad-Odyssey ProtocolNetFlownot the IANA assignmentThe conventional NetFlow collector port; IANA assigns it to Iliad-Odyssey.
2152tcp/udpgtp-userGTP-User Plane (3GPP)GTP user plane
3784tcp/udpbfd-controlBFD Control ProtocolBFD control
4739tcp/udp/sctpipfixIP Flow Info ExportIPFIX
4789udpvxlanVirtual eXtensible Local Area Network (VXLAN)VXLAN
6081udpgeneveGeneric Network Virtualization Encapsulation (Geneve)Geneve
6343tcp/udpsflowsFlow traffic monitoringsFlow
6653tcp/udpopenflowOpenFlowOpenFlow

VPN and tunnelling6

portprotoserviceused for
500tcp/udpisakmpisakmpIKE / ISAKMP
1194tcp/udpopenvpnOpenVPNOpenVPN
1701tcp/udpl2tpl2tpL2TP
1723tcp/udppptppptpPPTPUnencrypted. PPTP is cryptographically broken. Use IPsec or WireGuard.
4500tcp/udpipsec-nat-tIPsec NAT-TraversalIPsec NAT traversal
51820udpnot registeredWireGuardThe WireGuard default. Not registered with IANA.

Messaging and queues9

portprotoserviceused for
1883tcp/udpmqttMessage Queuing Telemetry Transport ProtocolMQTTUnencrypted. Use 8883 for MQTT over TLS.
4369tcp/udpepmdErlang Port Mapper DaemonErlang EPMD
5222tcpxmpp-clientXMPP Client ConnectionXMPP client
5269tcpxmpp-serverXMPP Server ConnectionXMPP server
5671tcp/udpamqpsamqp protocol over TLS/SSLAMQP over TLS
5672tcp/udp/sctpamqpAMQPAMQP / RabbitMQ
6667tcpircuIRCUIRC
8883tcp/udpsecure-mqttSecure MQTTMQTT over TLS
9092tcp/udpXmlIpcRegSvcXml-Ipc Server RegApache Kafkanot the IANA assignmentThe Kafka default; IANA assigns it to Xml-Ipc.

Containers and orchestration6

portprotoserviceused for
2181tcp/udpeforwardeforwardApache ZooKeepernot the IANA assignmentThe ZooKeeper default; IANA assigns it to eforward.
2375tcpdockerDocker REST API (plain text)Docker APIUnencrypted. Unauthenticated root-equivalent access. Never expose.
2376tcpdocker-sDocker REST API (ssl)Docker API over TLS
6443tcp/udpsun-sr-httpsService Registry Default HTTPS DomainKubernetes API servernot the IANA assignmentThe kube-apiserver default; IANA assigns it to Sun Service Registry.
10050tcp/udpzabbix-agentZabbix AgentZabbix agent
10051tcp/udpzabbix-trapperZabbix TrapperZabbix trapper

Voice and media5

portprotoserviceused for
554tcp/udprtspReal Time Streaming Protocol (RTSP)RTSP
3478tcp/udpstunSession Traversal Utilities for NAT (STUN) portSTUN / TURN
5060tcp/udp/sctpsipSIPSIPUnencrypted. Use 5061 for SIP over TLS.
5061tcp/udp/sctpsipsSIP-TLSSIP over TLS
32400tcpplexPlex multimediaPlex

Industrial and SCADA5

portprotoserviceused for
102tcp/udpiso-tsapISO-TSAP Class 0S7comm / ISO-TSAP
502tcp/udpmbapModbus Application ProtocolModbus TCPUnencrypted. No authentication in the protocol. Segment it.
2222tcp/udpEtherNet-IP-1EtherNet/IP I/O IANA assigned this well-formed service name as a replacement for "EtherNet/IP-1".EtherNet/IP implicit
20000tcp/udp/sctpdnpDNPDNP3
44818tcp/udpEtherNet-IP-2EtherNet/IP messaging IANA assigned this well-formed service name as a replacement for "EtherNet/IP-2".EtherNet/IP explicit

Legacy and diagnostics9

portprotoserviceused for
7tcp/udpechoEchoEcho
13tcp/udpdaytimeDaytimeDaytime
19tcp/udpchargenCharacter GeneratorChargenReflection-attack amplifier. Disable it.
37tcp/udptimeTimeTime
43tcp/udpnicnameWho IsWHOIS
79tcp/udpfingerFingerFinger
119tcp/udpnntpNetwork News Transfer ProtocolNNTP
194tcp/udpircInternet Relay Chat ProtocolIRC (original)
515tcp/udpprinterspoolerLPD print

Service names and descriptions come from the IANA Service Name and Transport Protocol Port Number Registry. Where the name in common use is not what IANA assigns, both are shown.

Also for subnetting

Try 5432, rdp, ms-wbt-server or 100.64. Arrow keys to move, Enter to open.

Subnet Calculator: IPv4 & IPv6 CIDR networking made simple

Enter a block in CIDR notation, IPv6 shorthand, or classic IP-and-mask form. Back comes the network address, the broadcast address, the usable host range, the subnet and wildcard masks, the host count and the address type, all recalculated the moment you type. It copes with a certification exercise and a data-centre addressing plan equally well, and it runs on a phone in a wiring closet as readily as on a desktop. No spreadsheets, no manual binary math.

Bulk subnet calculator: many networks at once, with CSV export

Most subnet calculators take one network at a time. This one treats a whole list as the normal case: enter one subnet per line (mixing IPv4 and IPv6 freely) and every line is calculated simultaneously, each with its network, broadcast, host range, masks and host count. Paste an entire addressing plan, a router configuration's worth of routes, or a firewall rule set, and the bulk subnet calculation happens as fast as you can paste. When you're done, export the complete result table to CSV for a spreadsheet or JSON for a script, or copy a share link that reproduces the exact list for a colleague. For network audits, IP address management and documentation, that batch workflow saves the hour that one-at-a-time calculators quietly cost.

An IPv4 and IPv6 subnet calculator in one

Most tools stop at version 4. This one handles the full range of legacy classful and modern classless IPv4 addressing, and the IPv6 side uses exact 128-bit arithmetic, so even a sprawling /48 or /64 returns a precise address count rather than an approximation. You can mix both families in the same list: each line is detected automatically, so there is no mode to switch. On a dual-stack network that matters more than it sounds, because the two halves of an addressing plan stop living in separate windows.

CIDR, netmask and mask conversion

Because modern routing is classless, the engine treats prefix length as the source of truth. Type 10.0.0.0/24, 192.168.1.1 255.255.255.0, or 2001:db8::/32 and it converts freely between prefix length, dotted-decimal netmask and wildcard mask. Confirming that a /26 is 255.255.255.192, or sizing a point-to-point link, takes as long as it takes to finish typing.

A visual, real-time subnet calculator

Every valid network becomes its own result row, with the address type colour-coded: private, public, loopback, link-local, documentation, carrier-grade NAT and the rest. Rows appear as you add subnets and expand to show masks, wildcard bits, total addresses and legacy class. That immediate feedback is what makes it usable in a classroom, where the point is seeing the boundary move, and in production, where the point is catching the typo before it reaches a router.

VLSM splitter and subnet summarization tool

Beyond single lookups, the app bundles a variable-length subnet mask planner and a subnet summarization tool. The VLSM splitter divides one parent block into right-sized child subnets from a simple list of host requirements, packing them largest-first so no address space is wasted and warning you if the parent block is too small. The aggregator acts as a route summarizer and supernet calculator, collapsing a long list of routes into the fewest CIDR blocks that cover exactly the same addresses. Handy for tidying route tables, access lists and firewall rules.

The calculator also converts in the direction people usually need it: CIDR to IP range. Enter 10.0.0.0/22 and you get the first and last address of the block, the usable host range between them, and the exact count — the same lookup an IP address calculator or a general network calculator is reached for, without switching tools for IPv6.

Subnet mask cheat sheet

The prefixes engineers reach for most, at a glance. Every value below comes straight from the calculator's own engine — and the full subnet cheat sheet has all of them, /0 to /32 for IPv4 and /0 to /128 for IPv6, with wildcard masks and a filter:

CIDR Subnet mask Wildcard Total IPs Usable hosts
/30255.255.255.2520.0.0.342
/29255.255.255.2480.0.0.786
/28255.255.255.2400.0.0.151614
/27255.255.255.2240.0.0.313230
/26255.255.255.1920.0.0.636462
/25255.255.255.1280.0.0.127128126
/24255.255.255.00.0.0.255256254
/23255.255.254.00.0.1.255512510
/22255.255.252.00.0.3.2551,0241,022
/16255.255.0.00.0.255.25565,53665,534
/8255.0.0.00.255.255.25516,777,21616,777,214

Learning to subnet, not just looking it up

A calculator answers the question; it does not make you faster at it, and in a CCNA or CompTIA Network+ exam you have about a minute per question and no calculator at all. The subnetting practice tool generates unlimited questions from this same engine — network and broadcast address, first and last usable host, host counts, masks and wildcards — marks them instantly and shows the working, so the arithmetic becomes something you do in your head rather than something you look up.

Why choose this online subnet calculator

Everything is computed in your browser, so the tool is quick, private, and carries on working offline once the page has loaded. Nothing you enter is sent to a server, there is no software to install and no account to create. A quick sanity check on a /24 and a full IPv6 addressing plan are both one keystroke away.

Accurate results you can trust

Under the hood, all subnet math is performed with exact integer arithmetic (including big-number support for IPv6) so host counts and network boundaries stay correct at every prefix length, from a /0 down to a single /32 or /128 host route. Copy any value with one click, export your results to CSV or JSON, or share a link that reloads the exact same list. However you work, this online subnet calculator keeps the numbers precise and the workflow quick, so it can be the last subnet reference you need to keep open. When the same troubleshooting session calls for it, the companion MAC address lookup resolves hardware vendors in bulk, the IP access list matcher tells you which ACL line permits or denies an address on Cisco, Arista or Juniper gear, and the data unit converter translates between bits, bytes, KB, MB and GB with the same exact-arithmetic guarantee.

Subnetting questions, answered

How do you calculate a subnet?

Start with an IP address and its prefix length (for example /24). The prefix says how many leading bits are the network portion; the remaining bits are host bits. Set every host bit to 0 to get the network address and to 1 to get the broadcast address. The addresses in between (minus those two on IPv4) are the usable host range.

From the prefix you also derive the subnet mask and the total number of addresses (2 raised to the number of host bits). A subnet calculator does all of this instantly: type a network in CIDR or IP-and-mask form and it returns the network, broadcast, host range, mask and host count.

What is 255.255.255.0 in subnet mask?

255.255.255.0 is the IPv4 subnet mask for a /24 network. In binary it is 24 ones followed by 8 zeros, so the first three octets identify the network and the last octet identifies the host.

A /24 holds 256 total addresses, of which 254 are usable for hosts. The first is the network address and the last is the broadcast address. It is the most common mask for home and small-office LANs, such as 192.168.1.0/24.

What is the subnet mask of 192.168.1.1 /24?

For 192.168.1.1/24 the subnet mask is 255.255.255.0 and the wildcard mask is 0.0.0.255. The /24 prefix means the first 24 bits are the network portion.

So the network address is 192.168.1.0 and the broadcast address is 192.168.1.255, leaving a usable host range of 192.168.1.1 to 192.168.1.254, 254 addresses in total.

How many hosts are in a /26 subnet mask?

A /26 subnet has 64 total addresses (26, since 32 − 26 = 6 host bits) and 62 usable host addresses after subtracting the network and broadcast addresses.

Its subnet mask is 255.255.255.192 and its wildcard mask is 0.0.0.63. A /26 is a common choice when you need to split a /24 into four equal subnets.

Can I calculate multiple subnets at once?

Yes: bulk subnet calculation is this calculator's normal mode. Enter one network per line, mixing IPv4 and IPv6 freely, and every line is calculated at the same time: network and broadcast address, usable host range, subnet and wildcard mask, and host count for each subnet in the list.

The whole result set exports to CSV or JSON in one click, and a share link reproduces the exact list: useful for documenting an addressing plan, auditing firewall rules, or checking a batch of routes without running them one at a time.

How to calculate an IPv6 subnet?

IPv6 subnetting uses the same prefix idea as IPv4, but with 128-bit addresses written in hexadecimal. Pick a prefix length such as /48 or /64: the bits before it are the network/subnet portion and the bits after it form the interface identifier. Set the interface bits to zero for the first address in the block.

Because IPv6 has no broadcast address and no reserved network or host addresses, every address in a prefix is usable. The number of addresses is 2 raised to (128 − prefix). A /64 (the standard size for a single LAN) holds 18,446,744,073,709,551,616 addresses.

What is the subnet in IPv6?

A subnet in IPv6 is a range of addresses that share a common prefix, identified by a prefix length after a slash (for example 2001:db8:abcd:1234::/64). The prefix is the network portion; the remaining bits form the interface identifier for individual hosts.

The universally recommended subnet size for a single link is /64. Larger allocations such as /48 or /56 are typically assigned to a whole site and then divided into many /64 subnets.

How many subnets are in IPv6?

It depends on the size of your allocation and the subnet size you choose. Each extra bit of prefix length doubles the number of subnets. Dividing a /48 into standard /64 subnets gives 264−48 = 65,536 subnets; a /56 into /64s gives 256 subnets.

In theory the entire IPv6 space split into /64s yields 264 subnets (about 18 quintillion) which is why IPv6 address exhaustion is not a practical concern.

What is my IPv6 subnet prefix length?

Your IPv6 subnet prefix length is the number after the slash in your address, such as the 64 in 2001:db8::1/64. It states how many leading bits are the network/subnet portion, and on most end devices and LANs it is /64.

To find it, check your router or interface configuration, or paste your IPv6 address into a subnet calculator. It reports the prefix length along with the network range and the exact address count.

What is a subnet calculator?

A subnet calculator is a tool that takes an IP network and instantly works out its addressing details, so you don't have to do binary math by hand. Enter a block in CIDR notation, IPv6 shorthand or IP-and-mask form and it returns the network address, broadcast address, usable host range, subnet mask, wildcard mask and host count for both IPv4 and IPv6.

It is used by network engineers, students and IT admins to plan networks, size subnets and verify configurations quickly and accurately.