Secure Document Search in Cloud Computing using MapReduce
Nowadays, cloud computing is an important technology, which is part of our daily lives. Moving to cloud brings some benefits: create new applications, store large sets of data, process large amount of data. Individual users or companies can store own data on cloud (e.g. maritime, environmental protection, physics analysis etc.). An important thing before storing in cloud is that data needs to be encrypted, in order to keep its confidentiality. Among these, users can store encrypted documents on cloud. However, when owner needs a specific document, they should retrieve all documents from cloud, decrypt them, chose the desired document, encrypt again and finally store back encrypted documents on cloud. To avoid these entire steps, a user can choose to work with searchable encryption. This is an encryption technique, where key words (or indexes) are associated to encrypted documents, and when the owner needs a document, he/she only needs to search throw key words and then retrieve the documents that have associated the desired keywords. An important programming paradigm for cloud computing is MapReduce, which allows high scalability on a large number of servers in a cluster. Basically, MapReduce works with (key, value) pairs. In the current study paper, we describe a new technique through which a user can extract encrypted documents stored on cloud servers based on key words, using searchable encryption and MapReduce.