Endianness

In the classic novel Gulliver's Travels by Jonathan Swift (1726), there is a silly debate between two groups of people on whether to break an egg through the wider end or the narrower end. These two groups of people are called big-endians and little-endians respectively.

# Computing

In today's computing landscape, endianness refers to the ordering of the bytes in a piece of data. Big-endian is an order that places the most significant value in the byte sequence first at the lowest storage address. Big-endian data is easier for human to read as the bytes are positioned from left to right.

Little-endian on the other hand places the least significant value first at the lower storage address. Little-endian data are non-intuitive for human to read, but it is easier for computers to perform computations and hence improves efficiency.

# Analogies and Examples

Here is a crude and simple analogy using the English sentence to represent in both big and little endian.

If you squint, it's basically reading the sentence from right to left for the sentence of little-endian.

A more realistic example is the representation of hexadecimal number.

This online converter tool can be used to visualize different endianness from a hexadecimal string.

# Little-endian Advantages

1. Incremental processing
2. Casts are no-op
3. Backward compatible
4. Ideal for devices with limited memory
5. Faster arithmetic operations such as addition and subtraction

The process of adding one to the number above is easy for little-endian as the least significant byte is at the start of the memory address whereas for big-endian, it requires traversal by the pointer to the least significant byte on the right before it can perform the addition.

# Extras

This is a function in C++ to find out whether our computer is Big-endian. The JVM is using Big-endian under the hood.

bool is_big_endian(void)
{
    union {
        uint32_t i;
        char c[4];
    } bint = {0x01020304};

    return bint.c[0] == 1;
}

# References