A formalized, taxonomy-driven approach to cross-layer application adaptation

Abstract

Advances in pervasive technology have made it possible to consider large-scale application types that potentially span heterogeneous organisations, technologies and device types. This class of application will have a multi-layer architecture, where each layer is likely to use languages and technologies appropriate to its own concerns. An example application is a geographically-large-scale crisis management system. Typically, such applications are required to dynamically adapt their behaviour based on current circumstances, with adaptations potentially affecting all layers of the application. The complexities involved in dynamically adapting multi-layer applications will significantly benefit from formal approaches to its specification. This paper presents a new methodology for flexible, multi-layer application adaptation, with layer-specific adaptation solution templates bound to application mismatches that are organised into hierarchical taxonomies. Templates can be linked either through direct invocations or through adaptation events, supporting flexible cross-layer adaptation. The methodology illustrates the use of different formalisms for different elements of its specification. In particular, we combine semi-formal metamodelling techniques for the system model specification with formal Petri nets, which are used to capture template matchmaking using reachability analysis. This work demonstrates how existing formalisms can be used for the specification of a generic adaptation model for pervasive applications.

A White Box Perspective on Behavioural Adaptation

“…Petri nets are used in [62] to formalize multi-layer adaptation in large scale applications spanning over heterogeneous organizations and technologies. Here the multi-layered architecture is motivated by the presence of different languages and technologies addressing their own concerns and views within the same application in a coherent manner and multi-layered adaptation must ensure that coherence between views is always maintained.…”

Model Driven Engineering of Cross-Layer Monitoring and Adaptation

“…We take a crisis-management system (CMS) (Popescu et al, 2012) as a sample multilayer system throughout this paper. When a flood incident is reported, an emergency centre performs rescue operations by organizing other departments to work together.…”

“…However, many constraints do not only depend on he current model state, but also the original state before the change. Taking the typical "missing role" mismatch as an example (Popescu et al, 2012). When we say we miss a role, we actually mean there was a role, but now, due to the system change, this role no longer exists.…”

“…The simulation had the similar function and structure as described by Popescu et al (Popescu et al, 2012). For the sake of simplicity, we implemented it based on the Spring platform 4 .…”

“…2) When an adaption is performed on one layer, how to identify all the complementary adaptions on other layers before executing them on the system. A typical idea towards cross layer system adaptation (Yuan et al, 2006;Zengin et al, 2011;Guinea et al, 2011;Popescu et al, 2012) is to regard the separated layers as a whole again, by explicitly defining the relations between the mismatches and solutions from different layers, and employ a centralized mechanism to handle them. These approaches actually violate a basic principle of multilayer systems, i.e., the separation of concerns between layers.…”