The Research On Outsourcing Computing For Large Matrix Multiplications In Cloud Computing

Cloud computing, as a new computing model, provides a new solution for computing and storage. Outsourcing computing has been proposed in accordance with the development of cloud computing: a computationally weak client is capable of delegating its heavy computing tasks, such as large matrix multiplications, to the cloud server, then the cloud server calculates the corresponding results and sends them back to the client. Obviously outsourcing computing not only improves the efficiency of computing but also avoids the waste of the computing resources.The emergence of outsourcing computing brings the new hope for the computationally weak client. Therefore there is important scientific significance and application value for researching on outsourcing computing. However, this new type of computing model has brought with it a series of problems which needed to be resolved as soon as possible. Firstly, secure outsourcing computing must ensure the result is unforgeblity: the results for a specific function may be incorrect as the cloud server may break down or be captured by an attacker. Therefore, the client must have the ability to check the correctness of the result. Also, secure outsourcing computing must preserve data security, including the privacy of the outsourced data and the privacy of the results. Note that the data usually contains some sensitive information of the client, thus a secure outsourcing computing scheme must prevent the exposure of data privacy to the third party. Finally, the efficiency of the verification must be achieved: in order to guarantee the feature of unforgeable results, the client needs to check the correctness of the results, so efficiency is the key point for outsourcing computing. Moreover, with the changing of application scenarios, the demand for the function of outsourcing computing is also growing. So far, the existing research of the outsourcing computing has the problem of inefficiency, functionality and security.Aimed at the general problems of outsourcing computing, two secure outsourcing computing schemes for the common scientific computing-large matrix multiplication are proposed in this paper.1. A private secure outsourcing computing for matrix multiplication scheme. In this work, the scheme we proposed can accomplish the outsourcing computing for the large matrix multiplication under the protection of the privacy for the outsourced data. In addition, the scheme also guarantees that the result is unforgeblity. Our work has a better efficiency, security and flexibility properties compared with the existing schemes through simulation results.2. A publicly verifiable secure matrix multiplication outsourcing scheme. In the proposed scheme, it achieves high efficiency, unforgeblity of the results and data privacy. Furthermore, it also achieves a new property named the public verification. Namely any third party can verify the results returned by cloud server. The security analysis demonstrates that our scheme is more secure and the performance evaluation indicates that this work is much more efficient compared with the existing works in terms of computation, communication and storage overhead.