Journal of Internet Services and Applications

This paper considers mission assurance for critical cloud applications, a set of applications with growing importance to governments and military organizations. Specifically, we consider applications in which assigned tasks or duties are performed in accordance with an intended purpose or plan in order to accomplish an assured mission. Mission-critical cloud computing may possibly involve hybrid (public, private, heterogeneous) clouds and require the realization of “end-to-end” and “cross-layered” security, dependability, and timeliness. We propose the properties and building blocks of a middleware for assured cloud computing that can support critical missions. In this approach, we assume that mission critical cloud computing must be designed with assurance in mind. In particular, the middleware in such systems must include sophisticated monitoring, assessment of policies, and response to manage the configuration and management of dynamic systems-of-systems with both trusted and partially trusted resources (data, sensors, networks, computers, etc.) and services sourced from multiple organizations.

The Internet of Things (IoT) is a promising concept toward pervasive computing as it may radically change the way people interact with the physical world, by connecting sensors to the Internet and, at a higher level, to the Web, thereby enacting a Web of Things (WoT). One of the challenges raised by the WoT is the in-network continuous processing of data streams presented by Things, which must be investigated urgently because it affects the future data models of the IoT, and is critical regarding the scalability and the sustainability required by the IoT. This cross-cutting concern has been previously studied in the context of Wireless Sensor Networks (WSN) given the focus on the acquisition and in-network processing of sensed data. However, proposed solutions feature various proprietary and highly specialized technologies that are difficult to integrate and complex to use, which represents a hurdle to their wide deployment. At the other end of the spectrum, cloud-based solutions introduce a too high energy cost for the envisioned IoT scale, considering the energy cost of communication over computation. There is thus a need for a distributed middleware solution for data stream management that leverages existing WSN work, while integrating it with today’s Web technologies in order to support the required flexibility and the interoperability of the IoT. Toward that goal, this paper introduces Dioptase, a lightweight Data Stream Management System for the WoT, which aims to integrate the Things and their streams into today’s Web by presenting sensors and actuators as Web services. The middleware specifically provides a way to describe complex fully-distributed stream-based mashups and to deploy them dynamically, at any time, as task graphs, over available Things of the network, including resource-constrained ones.

Highly complex distributed systems (HCDSs) are characterized by a large number of mission-critical, heterogeneous inter-dependent subsystems executing concurrently with diverse—often conflicting—quality-of-service (QoS) requirements. Creating, integrating, and assuring these properties in HCDSs is hard and expecting application developers to perform these activities without significant support is unrealistic. As a result, the computing and communication foundation for HCDSs is increasingly based on middleware. This article examines key R&D challenges that impede the ability of researchers and developers to manage HCDS software complexity. For each challenge that must be addressed to support HCDSs, the article surveys the state-of-the-art middleware solutions to these challenges and describes open issues and promising future research directions.

Vehicular communications are becoming a reality driven by various applications. Among those applications safe navigation support is of most significance. In designing such navigation safety applications, reliable dissemination of data, i.e., every affected vehicle receives data, is the key issue. Past research focused on the reliable dissemination problem of plain media type (e.g., text) safety messages, whereas we look at the problem of reliable and efficient dissemination of multimedia type (e.g., video, audio) safety information. Considering the potential volume of multimedia traffic in a large metropolis and the unpredictability of vehicular networks (e.g., high speed, partitions, obstacles, radio propagation anomalies, radio interference, etc.), reliable and efficient multimedia dissemination is non-trivial. By using a recently developed technique, network coding, we describe a method for reliable dissemination of video streams in case of emergencies. Simulation results show that in a typical setting, with representative channel errors/losses, our approach yields near 100% delivery ratio as compared to 92% delivery ratio by traditional multicasting. More importantly, the overhead is reduced by as much as 60%. Another important benefit is robustness to temporary disconnections. If the column of vehicles on the road has gaps, network coding jointly with “data muling” using vehicles in the opposite directions can deliver the multimedia files even to intermittently connected components.

In the Internet age, the concept of social network is emerging as both a useful means for understanding software development activities and a concept for designing software tools to support these activities. Employing Social Networking is useful in part, because it brings a focus to stakeholders in software development including developers, their managers, their support staff, QA analysts, requirement engineers, etc. Social networks represent how people communicate, coordinate, cooperate, and develop working relationships. The concept is also useful as it reflects the network-centric organization for dealing with highly inter-dependent artifacts–including complex source code, systems and subsystems, requirement and specification documents, etc. However, the knowledge on social network in software engineering is spread across the literature and the term social network is used for multiple purposes, such as social networking sites, social capital, interpersonal connections, and social structures. This thematic series puts together papers that employ the concept of social networks either in studying software development from an empirical approach or use the concept as a central basis for developing software tool support. Readers will find in this issue a centralized resource for research in this area.

A smart environment is equipped with numerous devices (i.e., sensors, actuators) that are possibly distributed over different locations (e.g., rooms of a smart building). These devices are automatically controlled to achieve different objectives related, for instance, to comfort, security and energy savings. Controlling smart environment devices is not an easy task. This is due to: the heterogeneity of devices, the inconsistencies that can result from communication errors or devices failure, and the conflicting decisions including those caused by environment dependencies. This paper proposes a design framework for the reliable and environment aware management of smart environment devices. The framework is based on the combination of the rule based middleware LINC and the automata based language Heptagon/BZR (H/BZR). It consists of: an abstraction layer for the heterogeneity of devices, a transactional execution mechanism to avoid inconsistencies and a controller that, based on a generic model of the environment, makes appropriate decisions and avoids conflicts. A case study with concrete devices, in the field of building automation, is presented to illustrate the framework.

In this paper, a distributed controller for a virtualized router is proposed. This controller enables the dynamic and automatic resource allocation between the different virtual routers (called slices) running on top of the physical router. The controller is designed on a two-layer architecture. A slice controller (one for each slice) estimates the relationship between the past performances and resource allocations of the slice using a linear model, and then determines the requested allocation for the slice to meet its target performance. The physical router consists of a set of modular linecards. A resource controller (one for each linecard), collects the resource allocation requests from the different slices using the resources it controls and determines the allocations based on the available capacities of the resources. Resources are allocated to slices to guarantee their target performances if possible, or provide service differentiation if the total requests from all the slices exceeds the capacities of the shared resources. We have found that the convergence of the controller depends on different parameters (such as the number of slices and the parameters of the linear model) and therefore some tuning of these parameters is needed for the system to achieve the stability.

Abstract Priority-based scheduling policies are commonly used to guarantee that requests submitted to the different service classes offered by cloud providers achieve the desired Quality of Service (QoS). However, the QoS delivered during resource contention periods may be unfair on certain requests. In particular, lower priority requests may have their resources preempted to accommodate resources associated with higher priority ones, even if the actual QoS delivered to the latter is above the desired level, while the former is underserved. Also, competing requests with the same priority may experience quite different QoS, since some of them may have their resources preempted, while others do not. In this paper we present a new scheduling policy that is driven by the QoS promised to individual requests. Benefits of using the QoS-driven policy are twofold: it maintains the QoS of each request as high as possible, considering their QoS targets and available resources; and it minimizes the variance of the QoS delivered to requests of the same class, promoting fairness. We used simulation experiments fed with traces from a production system to compare the QoS-driven policy with a state-of-the-practice priority-based one. In general, the QoS-driven policy delivers a better service than the priority-based one. Moreover, the equity of the QoS delivered to requests of the same class is much higher when the QoS-driven policy is used, particularly when not all requests get the promised QoS, which is the most important scenario. Finally, based on the current practice of large public cloud providers, our results show that penalties incurred by the priority-based scheduler in the scenarios studied can be, on average, as much as 193% higher than those incurred by the QoS-driven one.

With the Internet of Things (IoT), applications should interact with a huge number of devices and retrieve context data produced by those objects, which have to be discovered and selected a priori. Due to the number, heterogeneity, and dynamicity of resources, discovery services are required to consider many selection criteria, e.g., device capabilities, location, context data type, contextual situations, and quality. In this paper, we describe QoDisco, a semantic-based discovery service that addresses this requirement in IoT. QoDisco is composed of a set of repositories storing resource descriptions according to an ontology-based information model and it provides multi-attribute and range querying capabilities. We have evaluated different approaches to reduce the inherent cost of semantic search, namely parallel interactions with multiple repositories and publish-subscribe interactions. This paper also reports the results of some performance experiments on QoDisco with respect to these approaches to handle resource discovery requests in IoT.

Given the highly dynamic and extremely heterogeneous software systems composing the Future Internet, automatically achieving interoperability between software components —without modifying them— is more than simply desirable, it is quickly becoming a necessity. Although much work has been carried out on interoperability, existing solutions have not fully succeeded in keeping pace with the increasing complexity and heterogeneity of modern software, and meeting the demands of runtime support. On the one hand, solutions at the application layer synthesise intermediary entities, mediators, to compensate for the differences between the interfaces of components and coordinate their behaviours, while assuming the use of the same middleware solution. On the other hand, solutions at the middleware layer deal with interoperability across heterogeneous middleware technologies but do not reconcile the differences between components interfaces and behaviours at the application layer. In this paper we propose a unified approach for achieving interoperability between heterogeneous software components with compatible functionalities across the application and middleware layers. First, we provide a solution to automatically generate cross-layer parsers and composers that abstract network messages into a uniform representation independent of the middleware used. Second, these generated parsers and composers are integrated within a mediation framework to support the deployment of the mediators synthesised at the application layer. More specifically, the generated parser analyses the network messages received from one component and transforms them into a representation that can be understood by the application-level mediator. Then, the application-level mediator performs the necessary data conversion and behavioural coordination. Finally, the composer transforms the representation produced by the application-level mediator into network messages that can be sent to the other component. The resulting unified mediation framework reconciles the differences between software components from the application down to the middleware layers. We validate our approach through a case study in the area of conference management.