The problem of data security in cloud environment has been well studied. There exist numerous techniques to handle this issue. Some of the methods perform data encryption according to the profile of user, data nature etc. However, the security of such algorithms is not tampered proof and subject to compromise on guessing attacks. This affects the performance of such security algorithms. To handle this, a novel cryptographic scheme is presented in this article. The Adaptive QoS centric service orient dynamic cryptography (AQSDC) approach focuses on improving data security in the service orient environment. The method handles the client request and finds the data being accessed by the service request. Further, the service nature and the environmental and external network conditions are monitored and considered. According to the factors of various internal, external network conditions and service metrics, the method computes different crypto support measures like Crypto Traffic Class Support (CTCS), Crypto Throughput Class Support (CThCS), Crypto Tamper Class Support (CTaCS) and Crypto Data Class Support (CDCS). Using these measures, the method computes the sensitivity factor (SF) for the data being accessed. Accordingly, the method selects the class of algorithm to be used. Such algorithm from the selected class was used to split the service data into chunks and perform Class Centric Data Encryption (CCDE) algorithm which generates number of cipher blocks in a dynamic way. Generated blocks were reframed by the receiver for decryption. The proposed method improves the performance of data security with higher data encryption and decryption performance.