Optimal Data Placement and Replication Approach for SIoT with Edge

Social networks (SNs) are sources with extreme number of users around the world who all sharing data like images, audio, and video to their friends using IoT devices. This concept is so-called Internet Things (SIot). The evolving nature edge-cloud computing has enabled storage a large volume from various sources, this task demands an efficient procedure. For kind storage, usage replication edge geo-distributed cloud service area suited fulfill user’s expectations low latency. major issue way store replicate these items optimally allocate request center efficiently. data, we use server, which part in study. Thus, distributed stored quick access, will reduce latency response. proposed placement approach learns machine learning (ML) algorithm called radial basis kernel function assisted support vector (RBF-SVM) classify for storing user friend’s SIoT These algorithms be used predict workload as either or depending on existing time slots. dynamic also optimized graph partitioning (GP) method meet individual demand minimum costs. keep effective over time. Accordingly, introduces three kinds innovations compared approach. (i) Rather than single cloud, study uses server closest devices faster access reduced response (ii) classification RBF-SVM find reducing replication. (iii) Dynamic GP introduced cost fulfil SN. simulation result obtains 130 ms those approaches. Therefore, our ML-based provides promising results terms efficiency, effectiveness, performance cost.