Optimized IoT service placement in the fog

Abstract

The Internet of Things (IoT) leads to an evergrowing presence of ubiquitous networked computing devices in public, business, and private spaces. These devices do not simply act as sensors, but feature computational, storage, and networking resources. Being located at the edge of the network, these resources can be exploited to execute IoT applications in a distributed manner. This concept is known as fog computing. While the theoretical foundations of fog computing are already established, there is a lack of resource provisioning approaches to enable the exploitation of fogbased computational resources. To resolve this shortcoming, we present a conceptual fog computing framework. Then, we model the service placement problem for IoT applications over fog resources as an optimization problem, which explicitly considers the heterogeneity of applications and resources in terms of Quality of Service attributes. Finally, we propose a genetic algorithm as a problem resolution heuristic and show, through experiments, that the service execution can achieve a reduction of network communication delays when the genetic algorithm is used, and a better utilization of fog resources when the exact optimization method is applied.

How to place your apps in the fog: State of the art and open challenges

“…Unfortunately, none of the approaches previously discussed in this section released the code to run the experiments. Conversely, based on the hierarchical model of Skarlat et al, Venticinque and Amato proposed a software platform to support optimal application placement in the fog, within the framework of the CoSSMic European Project . Envisioning resource, bandwidth and response time constraints, their approach permits to choose among a cloud‐only, a fog‐only or a cloud‐to‐fog deployment policy, which were evaluated in a small testbed (ie, 1 Cloud, 1 fog node) where a composite application (eight components) from smart energy scenarios was deployed and run over emulated IoT data traces.…”

“…Skarlat et al designed a hierarchical modeling of fog infrastructures, consisting of a centralized management system to control fog nodes organized per colonies. [45][46][47] In particular, Skarlat et al 45 adopted an ILP formulation of the problem of allocating computation to fog nodes to optimize (user-defined) time deadlines on application execution, considering IoT devices needed to properly run the application. A simple linear model for cloud costs is also taken into account.…”

“…The resource capacity is usually modelled as a vector (or set) of elements, one for each of the hardware elements considered in the fog devices or the network. Examples of those vectors for the case of the fog devices are: considering CPU 24,58 ; considering CPU and storage 38 ; CPU, RAM, and storage 30,31,40,47 ; considering only storage. 39,41,45 Examples of resource capacity vectors both for fog nodes and network are: considering CPU, RAM, and network bandwidth 36,55 ; considering CPU, RAM, storage, and network bandwidth 33,50 ; considering processing and transmission capacities.…”

Quality of service‐aware approaches in fog computing

“…Moreover, Skarlat et al proposed a conceptual model for resource provisioning in FC. They also proposed a formal definition for the service placement problem in fog.…”

Distributed Fog‐to‐Cloud computing system: A minority game approach

“…Additionally, through the MG strategic interaction, the FN will be able to synchronize users' transmission tasks around q c to satisfy both the FN computing capacity and a changeable number of UNs transmission tasks yield Equation gives a simple and flexible tool for FN to control user transmission tasks around definite equilibrium level of N max , for a particular value of T l and UN QoE index β . In fact, by allowing UNs to run their tasks in the form of virtual machines (VMs), the SaaS model conforms with the “pay‐as‐you‐go” model of FNs and cloud as in …”