Vast collection of data, which is stored on the virtual storage, which we called as a Cloud. To protect this data from the unauthorized user cloud provider should implement all possible ways to make the cloud attack proof. In this paper, we have proposed a technique called (DROP) Methodology, Division, and Replication of Optimal Performance and Security that will make our system more secure from attacker to attack the system. In our System, once the data is being uploaded to the cloud the file will get divided into small fragments at multiple locations using T-colouring graph method which will store the small data at certain distance and at the random location so that attack will not be able to guess the location that will prohibit him to hack the data. In the system, we will also perform Encryption for each Fragment so that if attacker hacks anyone fragments he will able to only hack small amount of data which will useless for him as each fragment will have separate encryption key to prevent an attacker to hack it. After verifying the authentic user the fast Decryption process called Iris algorithm will be used which has high speed to collect all the fragments of provided data from the sender and true data will be received to the receiver.