Abstract:
When comparing conventional distributed computing like the grid and clustering, cloud computing is more actives with e- cient and low scienti c work ow costs. Every scienti c work requires massive data sets located in distinct data centers, resulting in a signi cant transmission delay. There are many challenges in cloud computing like security, routing, data placement, and Loading delay. Data placement and of Loading is a severe issue in the cloud; it impacts the execution time of data, performance, and cost. Cloud data centers are in remote locations; when data is accessed, and the process is executed, there comes enough delay. With the increment in the demands of the applications, the Internet of things (IoT) devices are expected to ful ll computation-intensive and latency-sensitive sensing and computational tasks, which is a challenge for these devices with limited battery and computation capability. Edge computing emerges as a promising strategy to minimize the latency and increase the productivity of these IoT devices. This research investigates the problems that come during data o oading and data placement. This research optimized the performance of IoT devices while performing computation by proposing an e cient strategy, Distributed Data O oading, and Placement (DDOP) algorithm for optimized data placement and of Loading. In the proposed algorithm, the tasks are rst scheduled and then coloaded to the data centers based on the earliest task completion time, which minimizes the latency and interference by monitoring the dependencies among the sub-tasks of the application. The proposed algorithm overcomes these factors and optimizes the o oading and placement of data.
