PROJECT TITLE :

Cloud Computing Service: The Caseof Large Matrix Determinant Computation

ABSTRACT:

Cloud computing paradigm provides another and economical service for resource-constrained purchasers to perform giant-scale knowledge computation. Since massive matrix determinant computation (DC) is ubiquitous in the fields of science and engineering, a initial step is taken in this paper to design a protocol that permits clients to securely, verifiably, and efficiently outsource DC to a malicious cloud. The main idea to guard the privacy is employing some transformations on the original matrix to urge an encrypted matrix which is distributed to the cloud; and then reworking the result returned from the cloud to induce the proper determinant of the initial matrix. Afterwards, a randomized Monte Carlo verification algorithm with one-sided error is introduced, whose superiority in designing cheap result verification algorithm for secure outsourcing is well demonstrated. In addition, it's analytically shown that the proposed protocol simultaneously fulfills the goals of correctness, security, robust cheating resistance, and high-potency. Intensive theoretical analysis and experimental evaluation conjointly show its high-potency and immediate practicability. It's hoped that the proposed protocol will shed light-weight in coming up with different novel secure outsourcing protocols, and encourage powerful corporations and working groups to finish the programming of the demanded all-inclusive scientific computations outsourcing software system. It's believed that such software system can be profitable by means that of providing massive-scale scientific computation services for so many potential clients.