Abstract
With the increasing number of Internet of Things (IoT) devices connected to the internet, a platform is required to support the enormous amount of data they generate. Since cloud computing is far away from the connected IoT devices, applications that require low-latency, real-time interaction and high quality of service (QoS) may suffer network delay in using the Cloud. Consequently, the concept of edge computing has appeared to complement cloud services, working as an intermediate layer with computation capabilities between the Cloud and IoT devices, to overcome these limitations. Although edge computing is a promising enabler for issues related to latency sensitivity, its deployment produces new challenges. Therefore, this paper presents and discusses the main factors of service latency for IoT applications, considering the impact of application characteristics (e.g., computation and communication demands), as well as resource utilization and heterogeneity at the edge level. Thus, a number of simulation experiments were set up to evaluate the influence of these factors. The outcomes of this research can be used to understand the complex interactions between many factors that affect the overall service time for latency-sensitive applications. Additionally, several open challenges are highlighted, serving as potential directions for future research.