Memory Usage Of Different Data Types in javascript

Question

A question that has happened to me is that different Data type in javascript how many use of memory . for Example in C++ data type like int , char , float uses order 2 , 1 , 8 byte of memory . now data Type like Number , string , boolean , null , undefind and Objects , Arrays in javascript how many use of memory and what is ranges that accepted ? Accept my apologize because of my low English level!!!

Answer 1

Numbers are 8 bytes.

Found that in this w3schools page.

I searched around a bit more for other JavaScript primitive types, but it's surprisingly hard to find this information! I did find the following code though:

    ...
    if ( typeof value === 'boolean' ) {
        bytes += 4;
    }
    else if ( typeof value === 'string' ) {
        bytes += value.length * 2;
    }
    else if ( typeof value === 'number' ) {
        bytes += 8;
    }
    ...

Seems to indicate that a String is 2 bytes per character, and a boolean is 4 bytes.

Found that code here and here. The full code's actually used to get the rough size of an object.

Although, upon further reading, I found this interesting code by konijn on this page: Count byte length of string.

function getByteCount( s )
{
  var count = 0, stringLength = s.length, i;
  s = String( s || "" );
  for( i = 0 ; i < stringLength ; i++ )
  {
    var partCount = encodeURI( s[i] ).split("%").length;
    count += partCount==1?1:partCount-1;
  }
  return count;
}
getByteCount("i♥js"); // 6 bytes
getByteCount("abcd"); // 4 bytes

So it seems that the string's size in memory depends on the characters themselves. Although I am still trying to figure out why he set the count to 1 if it's 1, otherwise he took count-1 (in the for loop).

Will update post if I find anything else.

Answer 2

As of today, MDN Data Structures page gives some more info about it:

Number

According to the ECMAScript standard, there is only one number type: the double-precision 64-bit binary format IEEE 754 value

So that should amount to 8 bytes.

String

JavaScript's String type is used to represent textual data. It is a set of "elements" of 16-bit unsigned integer values.

So that should amount to 2 bytes per character.

Boolean

Boolean represents a logical entity and can have two values: true, and false.

Nothing more about that.

Answer 3

The memory representation of a value, in other words its size, is an implementation detail unless it is specified otherwise, and can be different depending on the underlying code of the runtime environment. Runtime is the application that runs the JavaScript code i.e Nodejs, Firefox Browser or Chrome Browser.

Let's say you writing your own runtime and you want to represent a boolean value. A boolean value requires at least one bit, so you can store it in the memory as a single bit.

This is true:

■

This is false:

□

But there is nothing stopping you to use 8 bits or one byte memory space. Here most significant bit holds the value, rest will be padded with 0:

□□□□□□□■

Or 64 bits:

□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□□■

Same thing applies to a string value. You can represent it using 8 bits or 64bit or even more. However language promises to operate on any Unicode value, but 8 bits may not be large enough to represent all Unicode values. That is where you find some addressing and adjustments in the specification. The string in JavaScript is treated as a sequence of UTF-16 code units and some character takes up 16 bits memory space and others 32 bits.

Code unit is a term borrowed from the Unicode specification. You may ask why not a character but a code unit, because character is one way to represent a specific textual data and there are other ways too. For example both é, \u00E9 and \u{E9} creates letter e with acute. In your JavaScript string, you can prefer either of those.

Math is tricky because at CPU level all calculations are done in binary, and binary does not translate to decimal without issues. Try running 0.1 + 0.1 == 0.3. So we need tighter control for accuracy. That is why specification is very specific on the number representation: the double-precision 64-bit binary format IEEE 754.

To sum up, when writing a JavaScript runtime, how you store a value in the memory is up to you, unless it is clearly specified in the spec. That is why some data types have lengthy discussions on how they should be represented and stored in the memory.

Runtime environments are generally written in lower level languages such as C, C++ or Rust and memory management is delegated to what those languages use.

If you are JavaScript programmer you don't need to concern yourself how memory is allocated and freed because runtime handles that part for us. The only possible issue you might have is memory leaks through circular references: Object A has a reference to object B, and object B has a reference to object A. There are tools in the developer console to detect such issues.