Self-adaptive Architecture Research Articles

A self-adaptive heterogeneous multi-core architecture for embedded real-time video object tracking

Sequential Monte Carlo (SMC) represents a principal statistical method for tracking objects in video sequences by on-line estimation of the state of a non-linear dynamic system. The performance of individual stages of the SMC algorithm is usually data-dependent, making the prediction of the performance of a real-time capable system difficult and often leading to grossly overestimated and inefficient system designs. Also, the considerable computational complexity is a major obstacle when implementing SMC methods on purely CPU-based resource constrained embedded systems. In contrast, heterogeneous multi-cores present a more suitable implementation platform. We use hybrid CPU/FPGA systems, as they can efficiently execute both the control-centric sequential as well as the data-parallel parts of an SMC application. However, even with hybrid CPU/FPGA platforms, determining the optimal HW/SW partitioning is challenging in general, and even impossible with a design time approach. Thus, we need self-adaptive architectures and system software layers that are able to react autonomously to varying workloads and changing input data while preserving real-time constraints and area efficiency. In this article, we present a video tracking application modeled on top of a framework for implementing SMC methods on CPU/FPGA-based systems such as modern platform FPGAs. Based on a multithreaded programming model, our framework allows for an easy design space exploration with respect to the HW/SW partitioning. Additionally, the application can adaptively switch between several partitionings during run-time to react to changing input data and performance requirements. Our system utilizes two variants of a add/remove self-adaptation technique for task partitioning inside this framework that achieve soft real-time behavior while trying to minimize the number of active cores. To evaluate its performance and area requirements, we demonstrate the application and the framework on a real-life video tracking case study and show that partial reconfiguration can be effectively and transparently used for realizing adaptive real-time HW/SW systems.

Self-adaptive communication architecture of wireless sensor network based on Ptolemy Ⅱ

Self-adaptive communication architecture of wireless sensor network based on Ptolemy Ⅱ

Applying Bigraph Theory to Self-Adaptive Software Architecture

Existing theories of mobile and concurrency calculi for dynamic software architecture can not provide powerful support for evolutionary properties of self-adaptive software.Under this circumstance,Bigraph is based on a graphical model of mobile computation that emphasizes both locality and connectivity and has a complete and extensible theory framework.Therefore,Bigraph can provide a sound concept and intuitive,pervasive expression for self-adaptive software architecture.This paper introduces Bigraphical theory and current research,describes self-adaptive software architecture formally,analyzes and verifies some properties of dynamic evolution,and then discusses some promising directions of formal self-adaptive software architecture.

Dynamic Self-Adapting Software Architecture for Replica Management in Grids

Single data replica management strategy can not meet the complexity and diverseness in data grid. This paper proposes a dynamic self-adapting replica management architecture (DSA-RM) driven by software architecture (SA). In DSA-RM, architecture descriptions are shared by the whole grid environment as a system kernel to schedule various replica management components that are adapted to various application and VO in data grid. By the abstraction of DSA-RM, the paper describes the formal definitions of replica management components and discusses the dynamic self-adapting regulation of component evolution and multiplication in dynamic environment. By these regulations, the DSA-RM's framework and algorithms of multiplication are designed. Experimental results show that the DSA-RM can effectively promote the performance of grid system. Finally, the project of applying the

Network-based virtual personal overlay networks using programmable virtual routers

This article explores the use and adaptation of well established VPN technologies to dynamically network at run time devices and services in personal networks. Virtual personal overlay networks are envisaged as an extension of the basic VPN concept to network PN nodes in a secure manner. User managed and provider provisioned-based overlays are analyzed in terms of scalability, management complexity, and security to select the most appropriate VPON deployment and management framework. The dynamic nature of PNs calls for highly flexible self-configuring and self-adaptive VPON architectures. Programmable edge routers and dynamic tunneling mechanisms can be combined to meet the requirements set by PNs. The virtual router model is used in the provider edge to ease the dynamic deployment and management of VPONs. Results of an implementation and a performance evaluation of the VR concept based on open software routers are reported.