Computational Entities Research Articles

Editorial: Special Section on CMP Architectures

CHIP multiprocessor (CMP) architectures are formed when multiple compute cores are integrated onto the same chip, forming a single, powerful, computational entity. Nearly every major high-performance processor manufacturer has at least two cores (dual-core) on the die, and their roadmaps are increasingly multicore, signaling that the era of big, monolithic uniprocessors has ended. This results from the fact that ever-larger uniprocessors do not scale well in power/performance, area/performance, or design complexity/performance. Continued performance scaling of these processors will thus be focused primarily on increasing multithreaded throughput. The rapid adoption of small-scale CMP platforms and the quest for high performance continues to accelerate the rate at which processor manufacturers are considering adding more cores on the die. Over the last decade, there has been significant progress in research and development in both academia and industry on CMP architecture and design for client and server platforms. And, while we have successfully entered the era of CMP, there are a significant set of challenges and opportunities that are yet to be investigated deeply. Some of the broad research areas being investigated include CMP architecture alternatives (for core, cache, interconnect, and memory), CMP design and technologies (process implications, new technologies like 3D-stacking, voltage/clock domain management, etc.), CMP performance evaluation (new simulation and modeling techniques, emerging applications and execution environments like virtualization), and novel CMP architectures and use cases (asymmetric or heterogeneous architectures, accelerators, etc.). There are many questions that are still to be answered for CMP architectures. Below, we list a few of the most compelling ones.

Malleable applications for scalable high performance computing

Iterative applications are known to run as slow as their slowest computational component. This paper introduces malleability, a new dynamic reconfiguration strategy to overcome this limitation. Malleability is the ability to dynamically change the data size and number of computational entities in an application. Malleability can be used by middleware to autonomously reconfigure an application in response to dynamic changes in resource availability in an architecture-aware manner, allowing applications to optimize the use of multiple processors and diverse memory hierarchies in heterogeneous environments. The modular Internet Operating System (IOS) was extended to reconfigure applications autonomously using malleability. Two different iterative applications were made malleable. The first is used in astronomical modeling, and representative of maximum-likelihood applications was made malleable in the SALSA programming language. The second models the diffusion of heat over a two dimensional object, and is representative of applications such as partial differential equations and some types of distributed simulations. Versions of the heat application were made malleable both in SALSA and MPI. Algorithms for concurrent data redistribution are given for each type of application. Results show that using malleability for reconfiguration is 10 to 100 times faster on the tested environments. The algorithms are also shown to be highly scalable with respect to the quantity of data involved. While previous work has shown the utility of dynamically reconfigurable applications using only computational component migration, malleability is shown to provide up to a 15% speedup over component migration alone on a dynamic cluster environment. This work is part of an ongoing research effort to enable applications to be highly reconfigurable and autonomously modifiable by middleware in order to efficiently utilize distributed environments. Grid computing environments are becoming increasingly heterogeneous and dynamic, placing new demands on applications' adaptive behavior. This work shows that malleability is a key aspect in enabling effective dynamic reconfiguration of iterative applications in these environments.

A Bayesian Model for Event-based Trust

The application scenarios envisioned for ‘global ubiquitous computing’ have unique requirements that are often incompatible with traditional security paradigms. One alternative currently being investigated is to support security decision-making by explicit representation of principals' trusting relationships, i.e., via systems for computational trust. We focus here on systems where trust in a computational entity is interpreted as the expectation of certain future behaviour based on behavioural patterns of the past, and concern ourselves with the foundations of such probabilistic systems. In particular, we aim at establishing formal probabilistic models for computational trust and their fundamental properties. In the paper we define a mathematical measure for quantitatively comparing the effectiveness of probabilistic computational trust systems in various environments. Using it, we compare some of the systems from the computational trust literature; the comparison is derived formally, rather than obtained via experimental simulation as traditionally done. With this foundation in place, we formalise a general notion of information about past behaviour, based on event structures. This yields a flexible trust model where the probability of complex protocol outcomes can be assessed.

A Distributed Intelligent Agent Platform for Genetic Optimization in CEM: Applications in a Quasi-Point Matching Method

The use of genetic algorithms in computational electromagnetics (CEM) has proved a fruitful, but resource demanding optimization method with numerous practical applications. This paper introduces a parallel distributed computing framework based on the intelligent agent technology that is capable of handling the genetic optimization for diverse CEM problems. The platform core component is the genetic search agent (GSA), a collaborative computational entity that communicates with its peers in order to carry out a genetic optimization scheme. The platform can interface with foreign codes transparently due to its flexible code loading mechanisms, specially designed for CEM applications. The framework is applied to the optimization of a quasi-point matching method with fictitious sources (QPM-FS). The analysis comprises two case studies, involving electromagnetic scattering by: i) a two-layered cylinder; and ii) a cylinder buried in a two-layered earth medium, a problem of practical use in subsurface imaging. The performance results indicate that the sophisticated distribution mechanisms achieve high speed-up, while physically intuitive conclusions are obtained from the genetic optimization of the developed QPM-FS method

Agents come to bits: Towards a constructive comprehensive taxonomy of economic entities

Mirowski [Mirowski, P., 2007. Markets come to bits: evolution, computation, and markomata in economic science. Journal of Economic Behavior and Organization 63, 209–242] argues for a constructive approach to economic modeling centered on markets as evolving computational entities. This essay counters that a broader constructive approach to economic modeling can and should be taken. The recent advent of powerful computer technologies supporting agent-based modeling (ABM) renders feasible the computational study of economies modeled as evolving systems of interacting agents. In ABM, an “agent” refers broadly to bundled data and behavioral methods representing an entity constituting part of a computationally constructed world. Examples of possible agent referents include individuals, social groupings, institutions (e.g., markets), biological entities such as crops, and physical entities such as transportation networks and weather. Consequently, ABM provides tremendous opportunities for economists and other social scientists to tailor the breadth and depth of the entities represented in their models to the application at hand. A simple ABM of a two-sector decentralized market economy is used for concrete illustration.

Meso comes to markets: Comment on ‘Markets come to bits’

Mirowski's research programme of markomata formally sets out to model ‘the diversity of market forms by making use of the mathematical theory of automata, and then to taxonomize and organize these entities by means of a theory of evolution’ (Mirowski, this issue). This suggests a new sort of economics in which it is markets that evolve not people, systems that evolve not agents, prosthesis that evolve not preferences, and rules that evolve not information. Economies evolve; people do not. This proposition is both obvious and true, making it an excellent foundation for an evolutionary social science. So, if we are to take Mirowski's program seriously as a general framework for economic analysis, which we think we certainly should, then it is important to recognize that although Mirowski's unified rubric of markomata elides agency and is instead based upon John von Neumann's theory of automata as a model for the study of social institutions, the separate lines of research into markets as computational entities are all still very much but a microeconomics. Markomata is a theory of the core microeconomic building blocks of an economic system as well as a theory of their evolutionary principles (i.e. a microeconomics) but it is not yet a theory of the whole economy. Yet, it is our view that Mirowski's markomata program can indeed be the basis for such a general theory when analytically expounded in terms of the new evolutionary economic framework of micro meso macro (Dopfer et al., 2004 and Dopfer, 2005) in which market automata (Mirowski and Somefun, 1998 and Potts, 2001) are abstracted as a special, central instance of generic meso rules (Dopfer and Potts, 2004). Which is to suggest that the Mirowski framework of markomata can be yet further developed

Path-Computation-Element-Based Architecture for Interdomain MPLS/GMPLS Traffic Engineering: Overview and Performance

The Path Computation Element Working Group at the Internet Engineering Task Force is chartered to specify a PCE-based architecture for the path computation of interdomain MPLS- and GMPLS-based traffic engineered label switched paths. In this architecture, path computation does not occur at the head-end LSR, but on another path computation entity that may not be physically located on the head-end LSR. This method is greatly different from the traditional per-domain approach to path computation. This article presents analysis and results that compare performance of the PCE architecture with the current state-of-the-art approach. Detailed simulations are undertaken on varied and realistic scenarios where preliminary results show several performance benefits from the deployment of PCE. To provide a complete overview of significant development taking place in this area, milestones and progress at the IETF PCE WG are also discussed.

Specification and Verification of Agent Interaction Protocols in a Logic-based System

A number of information systems can be described as a set of interacting entities, which must follow interaction protocols. These protocols determine the behaviour and the properties of the overall system, hence it is of the uttermost importance that the entities behave in a conformant manner. A typical case is that of multi-agent systems, composed of a plurality of agents without a centralized control. Compliance to protocols can be hardwired in agent programs; however, this requires that only certified agents interact. In open systems, composed of autonomous and heterogeneous entities whose internal structure is, in general, not accessible (open agent societies being, again, a prominent example) interaction protocols should be specified in terms of the observable behaviour, and compliance should be verified by an external entity. In this paper, we propose a Java-Prolog- system for verification of compliance of computational entities to protocols specified in a logic-based formalism ( Social Integrity Constraints ). We also show the application of the formalism and the system to the specification and verification of three different scenarios: two specifications show the feasibility of our approach in the context of Multi Agent Systems (FIPA Contract-Net Protocol and Semi-Open societies), while a third specification applies to the specification of a lower level protocol (Open-Connection phase of the TCP protocol).

The leap from free markets to autonomous markets

In the spirit of recent work by Philip Mirowski, this article examines the consequences of extending the concept of free markets to a setting where markets are viewed as evolving computational entities. A mechanized market system that can perform the computational tasks required to allocate resources without interference from either humans or other mechanisms is said to be autonomous. At the present time, no computer-based market system, including those used on financial exchanges, is truly autonomous because various external mechanisms exist to override transactions and suspend trading. This article examines the computational feasibility of making markets autonomous and argues that a properly designed autonomous market would be evolutionarily fit.

Inside the brain of a neuron

For many decades, neurons were considered to be the elementary computational units of the brain and were assumed to summate incoming signals and elicit action potentials only in response to suprathreshold stimuli. Although modelling studies predicted that single neurons constitute a much more powerful computational entity, able to perform an array of nonlinear calculations, this possibility was not explored experimentally until the discovery of active mechanisms in the dendrites of most neuron types. Here, we review several modelling studies that have addressed information processing in single neurons, starting with those characterizing the arithmetic of different dendritic components, to those tackling neuronal integration at the cell body and, finally, those analysing the computational abilities of the axon. We present modelling predictions along with supporting experimental data in an effort to highlight the significant contribution of modelling work to enhancing our understanding of single-neuron arithmetic.

Manufacturing execution systems for customized production

Mass customised manufacturing is a current tendency in many production sectors. In this scenario, the client details the desired product often using informatics means and respecting available options. This customisation context imposes systemic integration and co-operations between the concerned manufacturing entities to explore their capabilities, aiming at adaptability to the product heterogeneity. An integration element of the management system and systems related to the shop floor is the manufacturing execution system (MES). A conceivable way to allow MES supporting this customised e-manufacturing is the concept of smart-product, where each product “drives ” its own production, allowing a decoupling between production and order dispatching, as well as the consistency between physical and informational flows. A smart-product requests services from manufacturing resources and it can compete for them. These resources must co-operate based on their features and based on some established flexible co-operation logic, aiming to carry out smart-products requests. However, this is by itself another issue, firstly due to the heterogeneity of factory resources. Looking for homogenisation and integration of this diversity, resources and also smart-products may be “encapsulated” inside of communicating entities called “holons ”. However, the composition of holonic MES (H-MES) is not trivial, because of the dynamic between all holons may be complex. Previous studies presented the organisation of resource-HLs co-operation carried out by computational entities called “rules”. In this paper, it is proposed a conceptual solution of a meta-model for rules-oriented and product-driven H-MES and its application for the holonification of a design and simulation tool.

The eightfold way of deliberation dialogue

Deliberation dialogues occur when two or more participants seek to jointly agree on an action or a course of action in some situation. We present the first formal framework for such dialogues, grounding it in a theory of deliberative reasoning from the philosophy of argumentation. We further fully articulate the locutions and rules of a formal dialogue game for this model, so as to specify a protocol for deliberation dialogues. The resulting protocol is suitable for dialogues between computational entities, such as autonomous software agents. To assess our protocol, we consider it against various records of human deliberations, against normative principles for the conduct of human dialogues, and with respect to the outcomes produced by dialogues undertaken according to the protocol. © 2007 Wiley Periodicals, Inc. Int J Int Syst 22: 95–132, 2007.

Combining Rate-Adaptive Cardiac Pacing Algorithms Via Multiagent Negotiation

Simulating and controlling physiological phenomena are notoriously complex tasks to tackle and require accurate models of the phenomena of interest. Currently, most physiological processes are described by a set of partial models capturing specific aspects of the phenomena, and usually their composition does not produce effective comprehensive models. A current open issue is thus the development of techniques able to effectively describe a phenomenon starting from partial models. This is particularly relevant for heart rate regulation modeling where a large number of heterogeneous partial models exists. In this paper we make the original proposal of adopting a multiagent paradigm, called anthropic agency, to provide a powerful and flexible tool for combining partial models of heart rate regulation for adaptive cardiac pacing applications. The partial models are embedded in autonomous computational entities, called agents, that cooperatively negotiate in order to smooth their conflicts on the values of the variables forming the global model the multiagent system provides. We experimentally evaluate our approach and we analyze its properties.

Biotope: An Integrated Framework for Simulating Distributed Multiagent Computational Systems

The study of distributed computational systems issues, such as heterogeneity, concurrency, control, and coordination, has yielded a number of models and architectures, which aspire to provide satisfying solutions to each of the above problems. One of the most intriguing and complex classes of distributed systems are computational ecosystems, which add an ecological perspective to these issues and introduce the characteristic of self-organization. Extending previous research work on self-organizing communities, we have developed Biotope, which is an agent simulation framework, where each one of its members is dynamic and self-maintaining. The system provides a highly configurable interface for modeling various environments as well as the living or computational entities that reside in them, while it introduces a series of tools for monitoring system evolution. Classifier systems and genetic algorithms have been employed for agent learning, while the dispersal distance theory has been adopted for agent replication. The framework has been used for the development of a characteristic demonstrator, where Biotope agents are engaged in well-known vital activities-nutrition, communication, growth, death-directed toward their own self-replication, just like in natural environments. This paper presents an analytical overview of the work conducted and concludes with a methodology for simulating distributed multiagent computational systems.

A multi-agent architecture for control of AGV systems

Agent is an autonomous, computational entity that can be viewed as perceiving its environment and acting upon it. Agents are event-driven objects that can be integrated in automated manufacturing environments to control certain tasks. In this paper a set of agents (a multi-agent system) is introduced to control an automated manufacturing environment. The architecture includes functions at the manufacturing cell level, materials handling and transport level, and factory scheduling level. Communication between these agents is accomplished by using a relational database (blackboard system). The relational database also integrates the requirements of a manufacturing execution system within the multi-agent task structure, which is unique to this architecture. Manufacturing cell and scheduling agents have been previously described in the literature. Here we focus our attention on the functions of the agents of the transport system, which is composed of a set of AGVs.

PumaMart: a parallel and autonomous agents based internet marketplace

Software agents are flexible, autonomous, and dynamic computational entities. For B2C e-commerce applications, the wide variety of choices to the consumers has also introduced the problem of information overloading. Meanwhile, there are so many e-shops and products for the consumers that it has become too time-consuming to find the best deal. In this paper, we present PumaMart, a Parallel and a utono mous agents based Internet Marke tplace, which deploys several novel models to facilitate autonomous and automatic online buying and selling by software agents (stationary and mobile) while providing fast response to consumers. These techniques include a 2-phase evaluation model, a parallel dispatch model and an auction-like negotiation model. Both evaluation model and negotiation model are based on the fuzzy evaluation criterion with clustering based grading function. What a consumer needs to do is to submit requests including the information for the desired products, selection preferences, through a web page in a Java-enabled browser. These information will be sent to the master agent at the server of the Agent Service Provider (ASP), which will employ its worker agents for subsequent shop searching/filtering, offer gathering/evaluating, negotiating, even booking and payment.

Agent oriented software engineering with web applications

This paper proposes an agent oriented approach to design web applications. Contrary to other research work, this paper considers agents only at the conceptual level because they are convenient abstractions. In the implementation, agents are substituted by more adequate computational entities like Enterprise Java Beans or servlets. The translation from the conceptual level to the implementation is based on the relevance of interactions in both web-applications and Multi-Agent Systems. Throughout the paper, we show the current situation in Multi-Agent systems and what elements are required in a Multi-Agent System implementation.

Embedded intelligent music–or iHiFi the intelligent HiFi

One consequence of current technological advances is that we can imbue everyday objects such as our fridges, ovens, fires, doors, thermostats, tvs and stereos with processing power. Not only will this enable them to make decisions based on their current state and model of the world about their current behaviour (the oven may automatically turn itself off when there is an excess of smoke) it will also allow communication with others to make decisions (the fridge may choose to defrost some vegetables when the phone communicates that there is a recent message from a friend who is, unexpectedly, intending to show for some dinner). This leads to the notion of ambient intelligence where intelligent computational entities are interwoven into the very fabric of our lives. We envisage a scenario where music devices cannot only make intelligent, sympathetic decisions about sound generation in order to satiate the particular sound requirements of a user, but would also interact with other devices both musical and otherwise in a massively dynamic and unpredictable environment. We also envisage that there will be a massive shift in the pattern of behaviour relating to music consumption; a move from the passive consumer to the active creator. The emerging field of computer science which is concerned with building systems which are inherently distributed, dynamic, open and social in this sense is known as intelligent agents. Whilst we are quite clear that we are in the very early tentative stages of this work, we nevertheless set out some realistic medium-term achievable functionalities for what we call intelligent embedded music, where intelligent, interactive music generation could continually and dynamically surround and sustain our day to day existence. We call this the intelligent HiFi (iHiFi).

ARKTOS: a knowledge engineering software tool for images

The goal of our ARKTOS project is to build an intelligent knowledge-based system to classify satellite sea ice images. It involves acquiring knowledge from sea ice experts, quantifying such knowledge as computational entities and ultimately building an intelligent classifier. In this paper we describe a two-stage knowledge engineering approach that facilitates explicit knowledge transfer, converting implicit visual cues and cognition of the experts to explicit attributes and rules implemented by the engineers. First, there is a prototyping stage that involves interviewing sea ice experts, transcribing the sessions, identifying descriptors and rules, designing and implementing the knowledge and delivering the prototype. The objective of this stage is to obtain a modestly accurate classification system quickly. Second, there is a refinement stage that involves evaluating the prototype, refining the knowledge base, modifying the design and re-evaluating the improved system. Since the refinement is evaluation-driven, the experts and the engineers are motivated explicitly to improve the knowledge base and are able to communicate with each other using a common, consistent platform. Moreover, since the classification result is immediately available, both sides are able to efficiently assess the correctness of the system. To facilitate the knowledge engineering of the second stage, we have designed and built three Java-based graphical user interfaces: arktosGUI, arktosViewer and arktosEditor. arktosGUI concentrates on feature-based refinement of specific attributes and rules. arktosViewer deals with regional evaluation. arktosEditor has a rule indexing and search mechanism and knowledge base editing capabilities.

Self-Organized Distributed Sensor Network Entity Tracking

The sensor network paradigm uses multiple autonomous sensor nodes to cooperatively construct an ad hoc computational entity. The authors have implemented applications using this approach with a network of sensor prototypes. In the paper, entity tracking is described as a two-tier process: ( i) clusters of nodes locally estimate parameters used in entity tracking (i.e. time, class, position, and heading), and ( ii) local parameter association forms inter-cluster tracks. Ad hoc routing primitives are used. This supports self-organization at all levels. Derivations of entity tracking applications are given along with preliminary results.