With the advent of cloud technology, the proliferation of cloud in various applications is enormous. Cloud can provide a variety of services. As major cloud service providers are public the data to be stored in cloud is at risk. A trivial solution would be keeping the files in encrypted form in the cloud, but whenever the file will be needed it will be downloaded and decrypted to get the data, but in a application which is user intensive this would lead to consumption of large bandwidth, hence it is not a feasible solution for providing a feasible computation and hence affecting the efficiency provided by the cloud. Therefore there is need for developing well planned solutions. In this work, we look into the problem of processing large amount of encrypted documents in XML formats. Here search or any computation operation can be performed on certain elements in the XML tree. The technique proposed makes use of index tables to allow fast addressing of keywords and location based queries. In order to perform computations on an untrusted server, symmetric encryption is used in along with homomorphic encryption to reduce computational and storage cost.