Dirk Bade

Context-Aware Computation Offloading for Mobile Cloud Computing: Requirements Analysis, Survey and Design Guideline

Abstract Along with the rise of mobile handheld devices the resource demands of respective applications grow as well. However, mobile devices are still and will always be limited related to performance (e.g., computation, storage and battery life), context adaptation (e.g., intermittent connectivity, scalability and heterogeneity) and security aspects. A prominent solution to overcome these limita- tions is the so-called computation offloading, which is the focus of mobile cloud computing (MCC). However, current approaches fail to address the complexity that results from quickly and constantly changing context conditions in mobile user scenarios and hence developing effective and efficient MCC applications is still challenging. Therefore, this paper first presents a list of re- quirements for MCC applications together with a survey and classification of current solutions. Furthermore, it provides a design guideline for the selection of suitable concepts for different classes of common cloud-augmented mobile applications. Finally, it presents open issues that developers and researchers should be aware of when designing their MCC-approach.

CloudAware: A Context-Adaptive Middleware for Mobile Edge and Cloud Computing Applications

The widespread use of mobile devices such as smartphones and tablets is accompanied by an ever increasing market for mobile applications, including resource demanding services like speech-or face recognition, that formerly were restricted to stationary devices. But as mobile devices remain comparatively limited in terms of resources (e.g., computation, storage and battery life), current approaches for augmentation have advocated the integration of cloud servers as well as other nearby devices to provide scalable computation-and storage resources to mobile end users. A current solution is the so-called computation offloading, which is the most prominent strategy used in Mobile Cloud Computing (MCC) and its successor known as Mobile Edge Computing (MEC). While MCC and MEC are receiving increasing attention, current work is often not able to cope with the quickly and constantly changing context (e.g., intermittent connectivity) of mobile devices. Therefore, this paper presents the evaluation of CloudAware, a context-adaptive mobile middleware for MCC as well as MEC that supports automated context adaptation by linking the distribution features of mobile middleware with context-aware self-adaptation techniques. In particular, we present a system software infrastructure and a data mining process which facilitate the development of elastic, scalable and context-adaptive mobile applications and present their evaluation using real usage data provided by the Nokia Mobile Data Challenge (MDC) dataset.

CloudAware: Empowering context-aware self-adaptation for mobile applications

Mobile devices are already woven into our everyday life, and we became accustomed that mobile applications assist us in a multitude of daily activities. With the rise of the Internet of Things, new opportunities to further automatize tedious tasks open up. New functional and user experience requirements demand for further resources and new ways to acquire these, because mobile devices remain comparatively limited in terms of, eg, computation, storage, and battery life. To face these challenges, current approaches augment mobile applications either with cloud resources (mobile cloud computing) or with resources near the mobile device at the logical edge of the network (mobile edge computing) onto which tasks can be offloaded during runtime. However, this does not automatically solve the conflict between resource demands and good user experience, as current solutions prove. It is the dynamically changing context that makes for good or bad offloading strategies. In this paper, we corroborate this finding by first evaluating 40 existing solutions based on a requirements catalogue derived from several application scenarios as well as the International Organization for Standardization/International Electrotechnical Commission criteria for software quality. Afterward, we present CloudAware, which is a mobile cloud computing/mobile edge computing middleware that supports automated context-aware self-adaptation techniques that ease the development of elastic, scalable, and context-adaptive mobile applications. Moreover, we present a qualitative evaluation of our concepts and quantitatively evaluate different offloading scenarios using real usage data to prove that mobile applications indeed benefit from context-aware self-adaptation techniques. Finally, we conclude with a discussion of open challenges.

An Agent-Based Event Processing Middleware for Sensor Networks and RFID Systems

Sensor networks as well as radio frequency identification (RFID) systems are among the hyped technologies nowadays. A lot of research efforts have been spent to develop standards, middleware and applications. The industry has already made large investments to foster the adoption of these technologies, consequently pushing the development, and already deploying the resulting technologies in different domains. However, the addressed technologies are still very young; best practices as well as standards are expected to frequently change, as new demands arise when using the technologies in everyday life. Because of this, middleware systems are expected to undergo frequent redesigns as well, requiring well-suited design paradigms to avoid a software engineering nightmare. We therefore propose an agent-based middleware for sensor networks and RFID systems. This middleware will meet the challenges for having a robust, adaptable and flexible middleware, which is, moreover, easily extensible to cope with expected re-engineerings and changes while maintaining a clear and elaborate design.

Generic context adaptation for mobile cloud computing environments

Markets for mobile applications offer myriads of apps ranging from simple to quite demanding ones. The latter are on the rise since every new generation of smartphones is equipped with more resources (CPU, memory, bandwidth, energy) to even allow re-source-demanding services like speech- or face recognition to be executed locally on a device. But compared to their stationary counterparts, mobile devices remain comparatively limited in terms of resources. Because of this, current approaches aim at extending mobile device capabilities with computation and storage resources offered by cloud services or other nearby devices. This paradigm, known as mobile cloud computing (MCC), is challenged by the dynamically changing context of mobile devices, which developers are required to take into account to decide, e.g., which application parts are when to offload. To rise to such and similar challenges we introduce the concept of Generic Context Adaptation (GCA), a data mining process that facilitates the adaptation of (mobile) applications to their current and future context. Moreover, we evaluate our approach with real usage data provided by the Nokia Mobile Data Challenge (MDC) as well as with CloudAware, a context-adaptive mobile middleware for MCC that supports automated and context-aware self-adaptation techniques.

From Smart Objects to Smarter Workflows -- An Architectural Approach

To live up to its promised business impact, the emerging Internet of Things must be accompanied by new ways to decentralize and adapt business logic for smart objects. Smart object lifecycle management refers to the decentralized, event-based, context-sensitive execution and adaptation of processes by smart physical objects during their path through different domains and their interaction with other entities. These entities include the end-user or consumer, other objects, or web-based information services. In this paper, we propose an approach how events and reaction to those events can be virtually attached to certain situations in the lifecycle of physical smart objects. In addition, we present an smart object lifecycle architecture for enabling multiple stakeholders to provide the necessary event and processing information for specific domains.

Generic Context Adaptation for Mobile Cloud Computing Environments

Abstract Markets for mobile applications offer myriads of apps ranging from simple to quite demanding ones. The latter are on the rise since every new generation of smartphones is equipped with more resources (CPU, memory, bandwidth, energy) to even allow resource-demanding services like speech- or face recognition to be executed locally on a device. But compared to their stationary counterparts, mobile devices remain comparatively limited in terms of resources. Because of this, current approaches aim at extending mobile device capabilities with computation and storage resources offered by cloud services or other nearby devices. This paradigm, known as Mobile Cloud Computing (MCC), is challenged by the dynamically changing context of mobile devices, which developers are required to take into account to decide, e.g., which application parts are when to offload. To rise to such and similar challenges we introduce the concept of Generic Context Adaptation (GCA), a data mining process that facilitates the adaptation of (mobile) applications to their current and future context. Moreover, we evaluate our approach with real usage data provided by the Nokia Mobile Data Challenge (MDC) as well as with CloudAware , a context-adaptive mobile middleware for MCC that supports automated context-aware self-adaptation techniques.

Integration von Kontextinformationen in Smart Applications und Smart Workflows

In Smart Mobile Applications sind virtuelle und physische Welt in immer weiter zunehmendem Mase miteinander verknupft: So werden u. a. insbesondere mobile Rechner und andere Gerate mit immer mehr Sensoren ausgestattet. Mit Hilfe derer konnen sie z. B. ihren eigenen Zustand, Attribute der physischen Umwelt, andere sie umgebende Objekte sowie Aktivitaten eines Benutzers wahrnehmen und fur neuartige Anwendungen auswerten. Hierfur bedarf es der Integration von Kontextinformationen in solche „smarten“ Anwendungen, was eine Reihe von Herausforderungen, insbesondere fur mobile Anwendungen, mit sich bringt. Im Rahmen dieses Artikels sollen daher verschiedene Moglichkeiten der Bereitstellung und existierende Ansatze der Integration betrachtet und hinsichtlich ihrer Eignung fur mobile Systeme diskutiert werden. Um die Integration und den Umgang mit Kontextinformationen weitergehend zu flexibilisieren, stellen wir zudem einen Top-down-Ansatz der Integration als konzeptionelle Erganzung zu existierenden Ansatzen sowie eine prototypische Systemunterstutzung vor.

An Awareness Model for Agents in Heterogeneous Environments

One of the constituting characteristics of software agents is their ability to sense the environment. The reception and processing of percepts is a key element for the agents internal reasoning process and essential for interacting with other entities in the environment. But sensing the environment is often seen as an abstract concept which is practically more or less reduced to the simple processing of some domain-specific message content. In order to be generally applicable among different multi-agent applications a common model of an environment incorporating an extensible set of entities, distribution protocols, and representation- as well as query languages needs to be established. Therefore, we propose a generic, extensible and adaptable model for resource-aware agents. It is organized into different information channels to help directing the focus of interest to specific aspects of the environment. Several discovery- and distribution protocols as well as different representation- and query languages may be used to satisfy the requirements of dynamic environments. The whole model is realized with a dedicated service agent on each platform, which local as well as remote agents can query for environmental information. This way, repeatedly and redundantly integrating these features into every agent application can be avoided and agent developers only have to deal with a simple protocol-API to access the information. Due to our highly flexible and adaptable model, we can face the heterogeneity of multi-agent applications operating in infrastructure- as well as mobile ad-hoc networks.

Abstract User Interfaces for Mobile Processes

An important focus of recent business process management systems is on the distributed, self-contained and even disconnected execution of processes involving mobile devices. Such an execution context leads to the class of mobile processes which are able to migrate between mobile and stationary devices in order to share functionalities and resources provided by the entire (mobile) environment. However, both the description and the execution of tasks which involve interactions of mobile users still require the executing device and its context to be known in advance in order to come up with a suitable user interface. Since this seems not appropriate for such decentralized and highly dynamic mobile processes, this work focuses on the integration of manual tasks on the respective ad-hoc creation of user interfaces at runtime. As an important prerequisite for that, this paper first presents an abstract and modality-independent interaction model to support the development and execution of user-centric mobile processes. Furthermore, the paper describes a prototype implementation for a corresponding system infrastructure component based on a service-oriented execution module, and, finally, shows its integration into the DEMAC (Distributed Environment for Mobility-Aware Computing) middleware.