Computing at the Edge Cloud: Performance Analysis of an Image Comparison Web Service

To improve the user's experience of Internet-based services (Web Services), Edge Cloud-based computing, in a variety of forms, is now recognized as a distinct entity within the Cloud computing paradigm. Serving nodes (e.g., computing or storage servers) are deployed at the edge of the Internet rather than at a distant, centralized data-centre to reduce latency and network traffic. Nevertheless, while it is acknowledged that such a deployment will result in performance improvements, few examples of quantified results exist in the literature. In this work, an image comparison Web Service has been developed and then deployed over both Edge Cloud and Core Cloud infrastructures. The performance of this Web Service has been studied by measuring data transfer time, processing time and total request and response time. The experimental results show that data transfer time and total request and response time in the Edge Cloud are reduced by more than 50% compared to the Core Cloud when the remote image database is situated in a separate server nearer to the user.