Abstract
Pervasive computing applications deal with the intelligence surrounding users that can facilitate their activities. This intelligence is provided in the form of software components incorporated in embedded systems or devices in close distance with end users. One example of infrastructure that can host intelligent pervasive services is the Edge Computing (EC) ecosystem. EC nodes can execute a number of tasks for data collected by devices present in the Internet of Things (IoT). In this paper, we propose an intelligent, proactive tasks management model based on demand. Demand depicts the number of users or applications interested in using the available tasks in EC nodes, thus characterizing their popularity. We rely on a Deep Machine Learning (DML) model and more specifically on a Long Short Term Memory (LSTM) network to learn the distribution of demand indicators for each task and estimate the future interest in them. This information is combined with historical observations of and support for a decision making scheme to conclude which tasks that are offloaded due to limited interest in them. We have to recognise that, in our decision making, we also take into consideration the load that every task may add to the processing node where it will be allocated. The description of our model is accompanied by a large set of experimental simulations for evaluating the proposed mechanism. We provide numerical results and reveal that the proposed scheme is capable of deciding on the fly, while concluding the most efficient decisions.
Highlights
The advent of the Internet of Things (IoT) offers many opportunities in the development of novel applications over a huge infrastructure of numerous devices
We propose a task management scheme where computation offloading is decided based on task demand; We adopt a Deep Machine Learning (DML)—i.e., Long Short Term Memory (LSTM)—model to estimate the future demand for each task present in an Edge Computing (EC) node; We provide an “aggregation” mechanism that combines past demand observations and future estimates to feed our reasoning mechanism and decide the tasks that should be offloaded to peers/Cloud; We support the “reasoning” mechanism of EC nodes adopting the principles of the multi-criteria theory; We provide an extensive experimental evaluation that reveals the pros and cons of the proposed approach
We focus on the discussed problem and take into consideration the dynamic nature of environments, such as the IoT or the EC where nodes interact
Summary
The advent of the Internet of Things (IoT) offers many opportunities in the development of novel applications over a huge infrastructure of numerous devices. Our “reasoning” is motivated by the demand that end users exhibit for every task It is a strategic decision for our model to incorporate the “popularity” of tasks in the selection process towards supporting popular tasks to be kept locally instead of being offloaded to other nodes. The intuition behind this is two fold: first, nodes save resources through the re-use of the tasks execution framework; secondly, the latency experienced by users is minimized as highly demanded tasks are initiated and executed immediately.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
