JoWUA

Abstract

There are several techniques to maximize the Quality of Service (QoS) of task scheduling in a cloud environment. The approaches can be classified according to the feature, metrics, and methods used in scheduling. For example, performance-based resource selection or service selection identifies the service per its throughput performance some of the ways involved in scheduling according to the makespan time, and idle time. Further, various metrics like resource availability, reputation, and popularity are used in scheduling tasks in a cloud environment. The above methods introduce poor service selection and scheduling performance because of missing the constraints like data fetch performance, data production performance, and performance of other devices. By considering all these, an efficient Location Aware DFS Scheduling Based QoS Maximization (LADFS-QM) algorithm is presented in this article. The method starts with preparing the scheduling data set for the environment by applying Resource Level Normalization algorithm, which involves removing noisy records and preparing the data accordingly. Second, Frequent Availability Optimizer (FAO) is used in resource selection, which computes Frequency and Availability Trust (FAT) value towards feature selection. Third, the method applies DFS scheduling in scheduling the cloud task, which estimates Data Fetch Support (DFS) measure, and ranks the services according to Data Contribution Support (DCS), which is being measured based on the number of IoT devices in route, number of reliable transmissions, number of retransmission and success rate. Finally, route selection is performed according to TFS (Trusted Forwarding Score), measured based on the IoT devices and their support on transmission, latency, congestion, etc. The proposed LADFS-QM algorithm improves the scheduling performance and the environment's quality of service.