Uma proposta para o uso de progressive web apps em ambientes de computação móvel em nuvens

Abstract

The increased use of mobile applications is directly associated with the constant development of Mobile Computing (CM), such as smartphones, tablets and notebooks. However, the use of applications on these devices turns out to be a problem for their processing limitations, RAM, storage, network connection and especially battery. To mitigate these limitations, an area known as Mobile Cloud Computing (MCC) has presented some alternatives. One of these is Computing Offloading, which consists of sending all or part of the execution of processes or tasks from the mobile device to a remote computing resource with higher processing power. However, the efficiency of this discharge from running in MCC environment can be directly related to the type of application used by the mobile device. Therefore, the objective of this work is to propose a methodology for the use of Progressive Web Applications or Progressive Web Apps (PWA) to offload tasks in Mobile Cloud Computing. Based on the proposed methodology, two case studies involving face detection and Fibonacci sequence calculation were developed. These case studies were evaluated by performing different workloads using the PWA application on different mobile devices and offloading in Docker containers using the 3G, 4G and WiFi networks. The results of the evaluation were obtained from the task execution time metric, noting that in some cases local execution of these tasks is better than cloud execution; while in others, cloud is a better processing option. The developed PWA application was further analyzed using the LightHouse PWA inspection tool, obtaining a maximum score of 100 for the application performance.