Optimizing Data Distribution And Privacy Protection Over Multi-Cloud Storage For Pdts Scheduling

Abstract

Privacy of electronic data that is of utmost priority is even in cloud computing environment. Putting data on a specific cloud service is quite risky since it may be leaked or compromised by an insider threat. Spreading some cloud storage providers (CSP) can be the solution to the preceding and will be of benefit by boosting the safety and file uploading by the dispersion of information on separate platforms. In such multi-cloud environments to reduce the time taken in uploading data, a suitable storage model was proposed. More intelligent erasure coding and task scheduling of partially dependent tasks (PDTs) forms the basis of this scheme and the task of this challenge will be perceived as a multi-objective optimization problem which falls in the category of NP-hard problems. A solution to this is provided by generating answers to this issue using a Non-dominated Sorting Genetic Algorithm (NSGA), which forms a collection of Pareto-optimal solutions of the conflicting objectives i.e. access time reduction, and increase in the level of data availability. When a user cannot make the best alternative that is proposed in Pareto front, entropy based decision-making process can be adopted to facilitate to make optimal decision. The fact of successful interaction between erasure coding and smart CSP selection matters to the privacy of the data and successful utilization of the multi-cloud storage. In order to enhance on this more the Muddy Soil Fish Optimization Algorithm (MSFOA) has been utilized together with a privacy-aware multi-objective optimization framework. This strategy together with a superior method of Pareto solution using NSGA transpires to outdo conventional security algorithms in several assessment aspects. The framework proposed is actually experimented in a variety of real life applications of the cloud storage and the same verifies the effectiveness of the framework.