Security analysis of unstructured data in NOSQL MongoDB database

NoSQL databases systems are non-relational databases uniquely intended to give high accessibility, reliability, and scalability for enormous data. Additionally sharding is the main fundamental favorable circumstances of NoSQL database. Various companies are moving towards NoSQL databases. NoSQL databases can store unstructured data such as email, multimedia, documents, and social media with high performance. NoSQL document stored database, MongoDB has many security risks which can be overcome by a good secure cryptographic system. In this paper, we will use symmetric cryptographic techniques for providing the security (confidentiality) of unstructured data in NoSQL document stored MongoDB. DES, AES, and blowfish algorithms with random key generation are used to encrypt/decrypt the document data before storing/retrieving to/from the NoSQL MongoDB database. We have also provided the comparative analysis of execution time taken by each algorithm with MongoDB for different size of data. There arises a problem that the storage size taken by the encrypted data in MongoDB database is more as compared to the original data. To solve this problem, we used a zlib compression technique to reduce the storage size taken by the encrypted data and provide comparative results.