TEMPOS: QoS Management Middleware for Edge Cloud Computing FaaS in the Internet of Things

Abstract

Several classes of advanced Internet of Things (IoT) applications, e.g., in the industrial manufacturing domain, call for Quality of Service (QoS) management to guarantee/control performance indicators, even in presence of many sources of ''stochastic noise'' in real deployment environments, from scarcely available bandwidth in a time window to concurrent usage of virtualized processing resources. This paper proposes a novel IoT-oriented middleware that i) considers and coordinates together different aspects of QoS monitoring, control, and management for different kinds of virtualized resources (from networking to processing) in a holistic way, and ii) specifically targets deployment environments where edge cloud resources are employed to enable the Serverless paradigm in the cloud continuum. The reported experimental results show how it is possible to achieve the desired QoS differentiation by coordinating heterogeneous mechanisms and technologies already available in the market. This demonstrates the feasibility of effective QoS-aware management of virtualized resources in the cloud-to-things continuum when considering a Serverless provisioning scenario, which is completely original in the related literature to the best of our knowledge.INDEX TERMS Edge cloud computing, FaaS, Internet of Things, interoperability, middleware, QoS management, serverless.

Resilient Edge: Building an Adaptive and Resilient Multi-Communication Network for IoT Edge Using LPWAN and WiFi

“…From the QoS perspective, multiple research papers have highlighted that the end-to-end perceived QoS on cloudedge continuum deployment environments depends on many complex system factors [53], [54], [55].…”

Optimization of Distributed System Energy Detection Method Considering Cloud Computing

“…In China, many enterprises have introduced the concept of cloud computing and conducted a lot of research. There are many solutions to the problem of power resource allocation in China, such as distributed generation access control strategy (GIS), renewable energy scheduling planning system based on load characteristics and time division management, and energy demand forecasting model based on geographic information system [7][8]. Therefore, based on cloud computing, this paper optimizes the energy detection methods of distributed systems.…”