Difference between JSON and bson

BSON
BSON By 10gen Developed a data format , At present, it is mainly used for MongoDB in , yes MongoDB Data storage format .BSON be based on JSON Format , choice JSON The main reason for the transformation is JSON The generality of and JSON Of schemaless Characteristics of .
By definition ：
BSON( Binary Serialized Document Format) Is a binary storage format , Similar to C The name of the language structure 、 For the representation , Support embedded document objects and array objects , Lightweight 、 Ergodicity 、 Characteristics of efficiency , It can effectively describe unstructured data and structured data .

BSON Is a kind JSON A storage format in binary form , abbreviation Binary JSON, It and JSON equally , Support embedded document objects and array objects , however BSON Yes JSON There are no data types , Such as Date and BinData type .
BSON It can be used as a storage form of network data exchange , This is a bit like Google Of Protocol Buffer, however BSON It's a kind of schema-less Storage form of , Its advantage is high flexibility , But its disadvantage is that the space utilization is not ideal .
in general ：
BSON There are three characteristics ： Lightweight 、 Ergodicity 、 Efficiency .

JSON
JSON yes JavaScript Object Notation An acronym for , The word means javascript Object notation , here json It means something similar to javascript A data format for an object .
Use JSON The format and parsing of can easily represent an object information ,json There are two formats ：① Object format ：{“key1”:obj,“key2”:obj,“key3”:obj…}、② Array / Set format ：[obj,obj,obj…].

1. Faster traversal speed
Yes json Format , Too big json Structure can cause data traversal to be very slow . stay json in , To skip a document for data reading , This document needs to be scanned , The need for cumbersome data structure matching , For example, the matching of parentheses .
　　 and BSON Yes json A big improvement of , It will be json The length of each element of exists in the head of the element , In this way, you only need to read the length of the element, and then you can seek Read at the specified point .

2. It's easier to operate
Yes JSON Come on , There is no type of data storage , For example, you need to modify the basic value , from 9 To 10, From one character to two , So maybe all the content behind it needs to be moved back one bit .
　　 While using BSON, You can specify this column as a number column , So no matter the number from 9 Grow to 10 still 100, We're all just making changes on the one where the numbers are stored , It will not cause the total length of data to increase .
　　 Of course , stay mongoDB in , If the number increases from plastic to long , The total length of data will increase .

3. Added additional data types
JSON Is a very convenient data exchange format , But its types are limited .
　　BSON On the basis of it, we add “byte array” data type . This eliminates the need for binary storage base64 After conversion, it will be saved as JSON, It greatly reduces the computation cost and data size .
　　 Of course , Sometimes ,BSON relative JSON There is no advantage in space , For example, yes. {“field”:7}, stay JSON On the storage of 7 Only one byte is used , And if used BSON, That is at least 4 Bytes （32 position ）

Currently in 10gen Under the efforts of ,BSON There are encoding and decoding packages for many languages . And are Apache 2 license Next open source . And still with mongoDB Further development .