Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Network byte order is big-endian: for a multi-byte number sent over a network, the most significant byte goes first. It gives protocols a shared wire format even when communicating computers use different byte orders in memory.
What does network byte order mean?
Network byte order is the standard ordering for multi-octet numeric values in Internet protocols. The most significant octet—the byte carrying the greatest place value—is transmitted first, followed by less significant octets. RFC 1700 states, “When a multi-octet quantity is transmitted the most significant octet is transmitted first.” (RFC 1700, Data Notations)
![Computer Networking Bible: [3 in 1] The Complete Crash Course to Effectively Design, Implement and Manage Networks. Including Sections on Security, Performance and Scalability](https://m.media-amazon.com/images/I/41H4YJnxKgL._SL160_.jpg)
This is a rule about the sequence of bytes in protocol data, not a requirement that every computer store numbers that way in memory. RFC 791 describes transmission order at the octet level, while NIST defines big-endian as a byte string ordered by decreasing significance, with the most significant byte at the left. (RFC 791, Appendix B; NIST glossary: big-endian)
Is network byte order big-endian or little-endian?
Network byte order is big-endian. Big-endian places the most significant byte first; little-endian places it last. These terms describe byte significance across a multi-byte value. The order of bits within an individual byte is a separate matter: RFC 1700 also identifies the leftmost bit as the most significant bit.
#1 Best Overall
| Convention | First byte of a multi-byte value | Where it applies |
|---|---|---|
| Network byte order | Most significant | Canonical representation specified by Internet protocols |
| Big-endian | Most significant | Byte-order convention; network byte order uses it |
| Little-endian | Least significant | A possible host-memory convention |
Why do protocols use a common byte order?
Different machines can use different conventions for laying out multi-byte values in memory. Without a defined wire order, the same sequence of bytes could be interpreted as different numbers on different systems. A protocol’s canonical order makes the representation predictable for both sender and receiver. The GNU C Library manual describes this purpose as enabling machines with different byte-order conventions to communicate. (GNU C Library manual, Byte Order Conversion)
What does network byte order look like?
Take the 32-bit value 0x12345678. Its network-order bytes are 0x12 0x34 0x56 0x78, from most significant to least significant. On a little-endian host, the same integer may appear in memory as 0x78 0x56 0x34 0x12. The value is unchanged; only the order of its bytes differs. The network sequence follows directly from the most-significant-octet-first rule.
How do you convert host byte order to network byte order in C?
In the GNU C Library’s socket guidance, use the conversion functions provided by netinet/in.h when the value must be represented in network order. The function name indicates the direction and the suffix indicates the width:
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errors| Function | Conversion | Width |
|---|---|---|
htons |
Host to network | 16 bits |
ntohs |
Network to host | 16 bits |
htonl |
Host to network | 32 bits |
ntohl |
Network to host | 32 bits |
For example, the GNU C Library manual says the sin_port and sin_addr fields in sockaddr_in must be in network byte order. Convert integer data encoded in socket messages as required by the protocol as well. The manual identifies htonl and ntohl as the pair used for IPv4 Internet addresses. (GNU C Library manual, Byte Order Conversion)
Conversion is about producing or interpreting the protocol’s representation; it is not necessarily a visible byte swap on every machine. On a host whose native order already matches network order, conversion can leave the bytes unchanged. Apply conversions at the serialization boundary, following the requirements of the protocol and API.
Quick Recap
Best Value
Rank #4
What network byte order does not mean
- It does not mean every host stores values in big-endian order in memory.
- It does not describe bit significance within a byte; that is a separate ordering question.
- It does not mean every numeric value in every application format should be converted the same way. Follow the wire format and API requirements for the field being serialized.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
