Compressing JSON using Data maps

Abstract: JSON (JavaScript Object Notation), is one of the widely used format for communication among software applications and devices. XML and CSV are other popular formats. One of the major advantages of JSON format is that it’s lightweight. There have been other formats which are starting to gain ground like Google’s Protocol Buffers, BSON etc. Additionally other representations of JSON have been researched. In this research paper we introduce a new representational format for JSON objects based on Data maps to further compress the size of a JSON data packet. We named it as FJSON where F stands for Flattened.We compare the size of JSON data packet with other formats i.e. XML, CJSON and FJSON. The use of Data maps eliminates the necessity of having delimiters as well as escaping special characters. Using examples, we show that there is a considerable impact on JSON data packet size in case of data packets having more key/value pairs and special characters. We also showcase that the amount of compression achieved using Data maps is directly proportional to the number of special characters or key-value pairs in a JSON data packet. Keywords: JSON, Data maps, Compression, Delimiters, Data Interchange Format, Protocol Buffer, BSON, CJSON