Search

Clustered architecture processors are preferred for embedded systems because centralized register file architectures scale poorly in terms of clock rate, chip area, and power consumption. Although clustering helps by improving clock speed, reducing energy consumption of the logic, and making the design simpler, it introduces extra overheads by way of inter-cluster communication. This communication happens over long global wires which leads to delay in execution and significantly high energy consumption. In this paper, we propose a new instruction scheduling algorithm that exploits scheduling slacks of instructions and communication slacks of data values together to achieve better energy-performance trade-offs for clustered architectures with heterogeneous interconnect. Our instruction scheduling algorithm achieves 35% and 40% reduction in communication energy, whereas the overall energy-delay product improves by 4.5% and 6.5% respectively for 2 cluster and 4 cluster machines with marginal increase (1.6% and 1.1%) in execution time. Our test bed uses the Trimaran compiler infrastructure. [ABSTRACT FROM AUTHOR]

One of the current trends in computer science leads to the design of computing organizations based on the activity of a multitude of tiny cheap decentralized computing entities. Whether these chips are integrated into paintings or disseminated in open environments like dust, the fundamental problem lies in their cooperative operation so that global functions are obtained collectively. In this paper, we address the issue of the creation of visual ambiences based on the coordinated activity of computing entities. These entities are distributed randomly on a 2D canvas and can only change their own color and perceive their immediate neighbors. [ABSTRACT FROM AUTHOR]

Motivated by trends in the industry towards transforming IT in large integrated service networks, this paper describes algorithms for the adaptive placement of "services" (as abstractions of collections of applications) in networks of "servers" (as abstractions for locations where services can be hosted). Networks comprised of interacting services as the foundation is also a vision pronounced by the Grid [9]. Manageability and "self-operation" of Grids is highly desirable. We analyze the requirements for algorithms one specific problem: the service placement problem. We discuss algorithms that neither require central control nor complete information about the system state. Algorithms are performed on a distributed overlay structure which summarizes load conditions in the underlying service network. The presented algorithms fulfill tasks of making initial placement decisions as well as initiating rearrangements when imbalance is detected. Presented algorithms have different characteristics regarding the tradeoff between accuracy (or quality) of a placement decision and its timeliness within which a decision can be made determining responsiveness. [ABSTRACT FROM AUTHOR]

Companies in today's automotive industry are under immense competitive pressure to reduce the length of their product development cycle from initial concept to begin of high-volume manufacturing. A very costly and immensely knowledge-intensive step in this process is the creation of tools and dies required to manufacture a car body of a specified design. This paper presents a novel architecture for a decision support system that streamlines the development process through the integration of a virtual assembly simulation, problem identification, and solution generation and evaluation. Following the virtual functional build process, our architecture deploys a number of multi-agent systems to provide system functionality, such as problem knowledge retrieval or solution generation and evaluation. [ABSTRACT FROM AUTHOR]

Textiles are omnipresent in everyday life. Their combination with microelectronics will lead to completely new applications, thus achieving elements of ambient intelligence. The integration of sensor or actuator networks, using fabrics with conductive fibres as a textile motherboard enable the fabrication of large active areas. In this paper we propose a "smart textile" based on a wired peer-to-peer network of simple information processing elements with integrated sensors or actuators. A self-organizing and fault-tolerant architecture is accomplished which detects the physical shape of the network. Routing paths are formed for data transmission, automatically circumventing defective or missing areas. The network architecture allows the smart textiles to be produced by reel-to-reel processes, cut into arbitrary shapes subsequently and implemented in systems at low installation costs. The possible applications are manifold, ranging from alarm systems to intelligent guidance systems, passenger recognition in car seats, air conditioning control in interior lining and smart wallpaper with software-defined light switches. [ABSTRACT FROM AUTHOR]

In this work, we are interested in modeling a production chain following the perspective of complex adaptive system. We propose an approach, allowing a production chain to manage by itself, its own behavior, so as to satisfy the constraints imposed upon it by its environment and reach a set of predefined objectives. We propose self-organization as a mechanism, to achieve such a goal. Keywords: self-organization, emergent behavior, swarm intelligence, software engineering. [ABSTRACT FROM AUTHOR]

Nowadays, lots of researches in Intrusion Detection and Intrusion Response try to find new solutions to circumvent new intrusive behaviors. One of the principal weaknesses of these systems is the lack of robustness inherent in their centralized nature. Even though most of the existing Intrusion Detection and Response Systems (IDRSystems) use distributed data collection (host-based or network-based) many of them continue to perform data analysis centrally, thereby limiting scalability. Moreover, even if the IDRSystem is distributed in the network, its deployed elements generally remain static. With the means available to modern attackers, such as automated intrusion tools, these static and distributed elements are easily accessible. Often, this does not always contribute to improving the reliability and resistance to attacks of such static components. This paper presents our approach for building an IDRSystem called Intrusion Detection and Response extended with Agent Mobility or IDReAM for short. IDReAM combines Mobile Agents (MAs) with self-organizing paradigms inspired by natural life systems. This approach was already announced in a preceding paper [4], and the present paper describes in a more detailed way the conceptual model. All the research works relating to IDReAM are gathered in a PhD Thesis [3] which also contains the implementation results of the model and its evaluation. The present paper is limited only to the model. Keywords: self-organization, mobile agents, intrusion detection and response. [ABSTRACT FROM AUTHOR]

This paper presents the design of new manufacturing coordination and control systems based on multi-agent technology. This design aims to cope with a dynamic environment characteristic for manufacturing systems nowadays. One important feature to handle these dynamics is having the ability to plan ahead, thus avoiding problems before they occur. Therefore, one novel characteristic of the system is the ability to perform emergent forecasting. Regarding emergent forecasting, an important issue that rises from this design is how to ensure that the forecast is reliable, and on the other hand, that the system is still fast enough to react against disturbances. This paper elaborates on the agents that form the system, and proposes a way to engineer it. Moreover, this paper also describes emergent forecasting. In addition to that, the trade off between responsiveness and forecast reliability (system nervousness issue) is also discussed in this paper, altogether with an example on the design of social acceptable behaviour that aims to handle the nervousness issue. Finally, some implementation and prototyping results are presented. [ABSTRACT FROM AUTHOR]

In recent years, mobile ad-hoc networks (MANET's) have been deployed in various scenarios, but their scalability is severely restricted by the human operators' ability to configure and manage the network in the face of rapid change of the network structure and demand patterns. In this paper, we present a self-organizing approach to MANET management based on stigmergic agents and demonstrate how to analyze its performance under different deployment assumptions. Our results emphasize the importance of attention to notions from dynamical systems theory in designing and deploying multi-agent systems. [ABSTRACT FROM AUTHOR]

Robotic soccer remains an area of active research owing to the difficulties of dynamic team formation and hard real time constraints regarding planning. Much of the existing research relies upon a central agency for coordination. Insect societies distribute work and allocate roles without a need for such a central agency and are robust with respect to changing environments and available agent resources. This paper explores the use of insect-inspired division of labour principles to robot soccer, highlighting the flexibility of the approach and ability to adapt to a wide range of soccer playing strategies. [ABSTRACT FROM AUTHOR]

Adaptation is an essential requirement for self-organizing multi-agent systems functioning in unknown dynamic environments. Adaptation allows agents to change their actions in response to environmental changes or actions of other agents in order to improve overall system performance, and remain robust even while a sizeable fraction of agents fails. In this paper we present and study a simple model of adaptation for task allocation problem in a multi-robot system. In our model robots have to choose between two types of task, and the goal is to achieve desired task division without any explicit communication between robots. Robots estimate the state of the environment from repeated local observations and decide what task to choose based on these observations. We model robots and observations as stochastic processes and study the dynamics of individual robots and the collective behavior. We validate our analysis with numerical simulations. [ABSTRACT FROM AUTHOR]

In this paper we present a general design methodology suitable for a class of complex multi-agent systems which are capable of self-assembly. Our methodology is based on a top-down, bottom-up approach, which has the potential to achieve a range of global design goals whilst retaining emergent behaviour somewhere in the system, and thereby allowing access to a richer solution space. Our experimental environment is a software system to model 2-dimensional self-assembly of groups of autonomous agents, where agents are defined as square smart blocks. The general design goal for such systems is to direct the self-assembly process to produce a specified structure. The potential of this design methodology has been realised by demonstrating its application to a "toy" problem - the self-assembly of rectangles of different sizes and shapes in a two-dimensional mesoblock environment. The design procedure shows different choices available for decomposing a system goal into subsidiary goals, as well as the steps needed to ensure a match to what is achievable from the bottom-up process. Encouraging results have been obtained, which allows mesoblock rectangles of specified size to be assembled in a directed fashion. Two different approaches to the same problem were presented, showing the flexibility of the method. [ABSTRACT FROM AUTHOR]

We present a programming approach to let a multitude of simple mobile computational "particles" (i.e. sorts of tiny mobile robots) to self-organize their respective locations to assume a coherent global formation (i.e. shape). The problem has a variety of applications in mobile robotics, modular robots, sensor networks, and computational self-assembly. Here we show how the TOTA ("Tuples On The Air") middleware can be effectively exploited to enable self-organization of spatial shapes in mobile particles with minimal capabilities. The key idea in TOTA is to rely on spatially distributed tuples, spread across the network, to drive particles' movements and activities. Several experiments are reported showing the effectiveness of the approach. [ABSTRACT FROM AUTHOR]

For Peer-2-Peer (P2P) networks to realize their full potential the nodes they are composed of need to coordinate and cooperate, to improve the performance of the network as a whole. This requires the suppression of selfish behavior (free-riding). Existing P2P systems often assume that nodes will behave altruistically, but this has been shown to be far from the case. We outline encouraging initial results from a P2P simulation that translates and applies the properties of "tag" models (initially developed within social simulations) [8, 9] to tackle these issues. We find that a simple node rewiring policy, based on the tag dynamics, quickly eliminates free-riding without centralized control. The process appears highly scalable and robust. [ABSTRACT FROM AUTHOR]

Contemporary software systems are exposed to demanding, dynamic, and unpredictable environments where the traditional adaptability mechanisms may not be sufficient. To imitate and fully benefit from life-like adaptability in software systems that might come closer to the complexity levels of biological organisms, we seek a formal mathematical model of certain fundamental concepts such as: life, organism, evolvability and adaptation. In this work we concentrate on the concept of software evolvability. Our work proposes an evolutionary computation model, based on the theory of hypercycles and autopoiesis. The intrinsic properties of hypercycles allow them to evolve into higher levels of complexity, analogous to multi-level, or hierarchical evolutionary processes. We aim to obtain structures of self-maintaining ensembles, that are hierarchically organised, and our primary focus is on such open-ended hierarchically organised evolution. [ABSTRACT FROM AUTHOR]

The ‘engineering' and ‘adaptive' approaches to system production are distinguished. It is argued that producing reliable self-organised software systems (SOSS) will necessarily involve considerable use of adaptive approaches. A class of apparently simple multi-agent systems is defined, which however has all the power of a Turing machine, and hence is beyond formal specification and design methods (in general). It is then shown that such systems can be evolved to perform simple tasks. This highlights how we may be faced with systems whose workings we have not wholly designed and hence that we will have to treat them more as natural science treat the systems it encounters, namely using the classic experimental method. An example is briefly discussed. A system for annotating such systems with hypotheses, and conditions of application is proposed that would be a natural extension of current methods of open source code development. [ABSTRACT FROM AUTHOR]

It is our belief that modelling the behaviour of stem cells in the adult human body as an agent-based system is the most appropriate way of understanding the process of self-organisation. We have undertaken several case studies where formal and/or computational models of stem cell systems, have been re-developed using an agent-based approach. This paper presents details of one of these case studies where we have used an agent-based approach as opposed to a cellular automata approach. A formalisation of the non-agent and agent-based approach is given, and from the results of this investigation, we aim to demonstrate the advantages of the agent-based approach for developing biologically plausible models with emergent self-organising dynamics. The aim of this paper first to discuss the importance of modelling and simulating stem cells, because of certain experimental limitations, but also to demonstrate that the multi-agent approach to modelling is the most appropriate. [ABSTRACT FROM AUTHOR]

Positioned at the confluence between human/machine and hardware/software integration and backed by a solid proof of concept realized through several scenarios encompassing e-Securities, e-Health, and e-Logistics for global manufacturing and emergency response management, this work exploits latest advances in information and networking technologies to set a systematic framework for the design of the information infrastructures (coined as AIIs - Adaptive Information Infrastructures) destined to fuel tomorrow's e-Society. Designed following the natural laws of evolution, which merge self-organization and natural selection [1], these socially embedded information infrastructures can adapt to fulfill various needs as their environment demands. Computational intelligence techniques endow the AIIs with learning and discovery capabilities, emulating social and biological behavior. AIIs are destined to become an integral part of our life by supporting, rather than disturbing, a framework that facilitates strategic partnerships while providing greater user-friendliness, more efficient services support, user-empowerment, and support for human interactions. Keywords. distributed artificial intelligence, information infrastructures, emergency response management, e-Health, Cybersecurities, emergence, self-organization, evolution [ABSTRACT FROM AUTHOR]

We describe an analytic approach, multiscale analysis, that can demonstrate the fundamental limitations of decomposition based engineering for the development of highly complex systems. The planning based process is limited by the interdependence of components and communication between design teams. Thus, the construction of many highly complex systems should be pursued by strategies modeled after biological evolution, or market economies, where extensive planning is forsaken and multiple parallel design efforts compete for adoption through testing in actual use. [ABSTRACT FROM AUTHOR]

A clear terminology is essential in every research discipline. In the context of ESOA, a lot of confusion exists about the meaning of the terms emergence and self-organisation. One of the sources of the confusion comes from the fact that a combination of both phenomena often occurs in dynamical systems. In this paper a historic overview of the use of each concept as well as a working definition, that is compatible with the historic and current meaning of the concepts, is given. Each definition is explained by supporting it with important characteristics found in the literature. We show that emergence and self-organisation each emphasise different properties of a system. Both phenomena can exist in isolation. The paper also outlines some examples of such systems and considers the combination of emergence and self-organisation as a promising approach in complex multi-agent systems. [ABSTRACT FROM AUTHOR]

Unending quest for performance improvement coupled with the advancements in integrated circuit technology have led to the development of new architectural paradigm. Speculative multithreaded architecture (SpMT) philosophy relies on aggressive speculative execution for improved performance. However, aggressive speculative execution comes with a mixed flavor of improving performance, when successful, and adversely affecting the performance (and energy consumption) because of useless computation in the event of mis-speculation. Dynamic instruction criticality information can be applied to control and guide such an aggressive speculative execution. In this paper, we propose a model to determine the dynamic instruction criticality of SpMT execution. We have also developed two novel techniques, utilizing the criticality information, namely delaying the non-critical loads and the criticality based thread-prediction for reducing useless computations. Our experiments with criticality based speculation control show a significant reduction in useless computation with little reduction in speedup. [ABSTRACT FROM AUTHOR]