Automating fault tolerance in high-performance computational biological jobs using multi-agent approaches

Background: Large-scale biological jobs on high-performance computing systems require manual intervention if one or more computing cores on which they execute fail. This places not only a cost on the maintenance of the job, but also a cost on the time taken for reinstating the job and the risk of losing data and execution accomplished by the job before it failed. Approaches which can proactively detect computing core failures and take action to relocate the computing core's job onto reliable cores can make a significant step towards automating fault tolerance. Method: This paper describes an experimental investigation into the use of multi-agent approaches for fault tolerance. Two approaches are studied, the first at the job level and the second at the core level. The approaches are investigated for single core failure scenarios that can occur in the execution of parallel reduction algorithms on computer clusters. A third approach is proposed that incorporates multi-agent technology both at the job and core level. Experiments are pursued in the context of genome searching, a popular computational biology application. Result: The key conclusion is that the approaches proposed are feasible for automating fault tolerance in high-performance computing systems with minimal human intervention. In a typical experiment in which the fault tolerance is studied, centralised and decentralised checkpointing approaches on an average add 90% to the actual time for executing the job. On the other hand, in the same experiment the multi-agent approaches add only 10% to the overall execution time

Design and Qualitative/Quantitative Analysis of Multi-Agent Spatial Simulation Library

Thesis (Master's)--University of Washington, 2012

A multi-agent-based evolution model of innovation networks in dynamic environments

An innovation network can be considered as a complex adaptive system with evolution affected by dynamic environments. This paper establishes a multi-agent-based evolution model of innovation networks under dynamic settings through computational and logical modeling, and a multi-agent system paradigm. This evolution model is composed of several sub-models of agents' knowledge production by independent innovations in dynamic situations, knowledge learning by cooperative innovations covering agents' heterogeneities, decision-making for innovation selections, and knowledge update considering decay factors. On the basis of above-mentioned sub-models, an evolution rule for multi-agent based innovation network system is given. The proposed evolution model can be utilized to simulate and analyze different scenarios of innovation networks in various dynamic environments and support decision-making for innovation network optimization.

DITRANS - a multi-agent system for integrated diagnosis of power transformers

This paper describes the development and the implementation of a multi-agent system for integrated diagnosis of power transformers. The system is divided in layers which contain a number of agents performing different functions. The social ability and cooperation between the agents lead to the final diagnosis and to other relevant conclusions through integrating various monitoring technologies, diagnostic methods and data sources, such as the dissolved gas analysis.

MASGriP – a Multi-Agent Smart Grid Simulation Platform

The increasing number of players that operate in power systems leads to a more complex management. In this paper a new multi-agent platform is proposed, which simulates the real operation of power system players. MASGriP – A Multi-Agent Smart Grid Simulation Platform is presented. Several consumer and producer agents are implemented and simulated, considering real characteristics and different goals and actuation strategies. Aggregator entities, such as Virtual Power Players and Curtailment Service Providers are also included. The integration of MASGriP agents in MASCEM (Multi-Agent System for Competitive Electricity Markets) simulator allows the simulation of technical and economical activities of several players. An energy resources management architecture used in microgrids is also explained.

Multi-agent simulation of continental, regional, and micro electricity markets

Renewable based power generation has significantly increased over the last years. However, this process has evolved separately from electricity markets, leading to an inadequacy of the present market models to cope with huge quantities of renewable energy resources, and to take full advantage of the presently existing and the increasing envisaged renewable based and distributed energy resources. This paper proposes the modelling of electricity markets at several levels (continental, regional and micro), taking into account the specific characteristics of the players and resources involved in each level and ensuring that the proposed models accommodate adequate business models able to support the contribution of all the resources in the system, from the largest to the smaller ones. The proposed market models are integrated in MASCEM (Multi- Agent Simulator of Competitive Electricity Markets), using the multi agent approach advantages for overcoming the current inadequacy and significant limitations of the presently existing electricity market simulators to deal with the complex electricity market models that must be adopted.

Multi-agent electricity market simulation with dynamic strategies & virtual power producers

Distributed energy resources will provide a significant amount of the electricity generation and will be a normal profitable business. In the new decentralized grid, customers will be among the many decentralized players and may even help to co-produce the required energy services such as demand-side management and load shedding. So, they will gain the opportunity to be more active market players. The aggregation of DG plants gives place to a new concept: the Virtual Power Producer (VPP). VPPs can reinforce the importance of these generation technologies making them valuable in electricity markets. In this paper we propose the improvement of MASCEM, a multi-agent simulation tool to study negotiations in electricity spot markets based on different market mechanisms and behavior strategies, in order to take account of decentralized players such as VPP.

A new approach for multi-agent coalition formation and management in the scope of electricity markets

This paper presents a new methodology for the creation and management of coalitions in Electricity Markets. This approach is tested using the multi-agent market simulator MASCEM, taking advantage of its ability to provide the means to model and simulate VPP (Virtual Power Producers). VPPs are represented as coalitions of agents, with the capability of negotiating both in the market, and internally, with their members, in order to combine and manage their individual specific characteristics and goals, with the strategy and objectives of the VPP itself. The new features include the development of particular individual facilitators to manage the communications amongst the members of each coalition independently from the rest of the simulation, and also the mechanisms for the classification of the agents that are candidates to join the coalition. In addition, a global study on the results of the Iberian Electricity Market is performed, to compare and analyze different approaches for defining consistent and adequate strategies to integrate into the agents of MASCEM. This, combined with the application of learning and prediction techniques provide the agents with the ability to learn and adapt themselves, by adjusting their actions to the continued evolving states of the world they are playing in.

Multi-agent based electricity market simulator with VPP: Conceptual and implementation issues

This paper presents a new architecture for the MASCEM, a multi-agent electricity market simulator. This is implemented in a Prolog which is integrated in the JAVA program by using the LPA Win-Prolog Intelligence Server (IS) provides a DLL interface between Win-Prolog and other applications. This paper mainly focus on the MASCEM ability to provide the means to model and simulate Virtual Power Producers (VPP). VPPs are represented as a coalition of agents, with specific characteristics and goals. VPPs can reinforce the importance of these generation technologies making them valuable in electricity markets.

Multi-apprentice learning for meta-heuristics parameter tuning in a multi agent scheduling system

The scheduling problem is considered in complexity theory as a NP-hard combinatorial optimization problem. Meta-heuristics proved to be very useful in the resolution of this class of problems. However, these techniques require parameter tuning which is a very hard task to perform. A Case-based Reasoning module is proposed in order to solve the parameter tuning problem in a Multi-Agent Scheduling System. A computational study is performed in order to evaluate the proposed CBR module performance.

Cooperation mechanism for team-work based multi-agent system in dynamic scheduling through meta-heuristics

This paper describes a Multi-agent Scheduling System that assumes the existence of several Machines Agents (which are decision-making entities) distributed inside the Manufacturing System that interact and cooperate with other agents in order to obtain optimal or near-optimal global performances. Agents have to manage their internal behaviors and their relationships with other agents via cooperative negotiation in accordance with business policies defined by the user manager. Some Multi Agent Systems (MAS) organizational aspects are considered. An original Cooperation Mechanism for a Team-work based Architecture is proposed to address dynamic scheduling using Meta-Heuristics.

Multi-agent system for personalization of location-based services

The current ubiquitous network access and increase in network bandwidth are driving the sales of mobile location-aware user devices and, consequently, the development of context-aware applications, namely location-based services. The goal of this project is to provide consumers of location-based services with a richer end-user experience by means of service composition, personalization, device adaptation and continuity of service. Our approach relies on a multi-agent system composed of proxy agents that act as mediators and providers of personalization meta-services, device adaptation and continuity of service for consumers of pre-existing location-based services. These proxy agents, which have Web services interfaces to ensure a high level of interoperability, perform service composition and take in consideration the preferences of the users, the limitations of the user devices, making the usage of different types of devices seamless for the end-user. To validate and evaluate the performance of this approach, use cases were defined, tests were conducted and results gathered which demonstrated that the initial goals were successfully fulfilled.

Environmental Decision Support: a Multi-Agent Approach

The environmental management domain is vast and encompasses many identifiable activities: impact assessment, planning, project evaluation, etc. In particular, this paper focusses on the modelling of the project evaluation activity. The environmental decision support system under development aims to provide assistance to project developers in the selection of adequate locations, guaranteeing the compliance with the applicable regulations and the existing development plans as well as satisfying the specified project requirements. The inherent multidisciplinarity features of this activity lead to the adoption of the Multi-Agent paradigm, and, in particular, to the modelling of the involved agencies as a community of cooperative autonomous agents, where each agency contributes with its share of problem solving to the final system’s recommendation. To achieve this behaviour the many conclusions of the individual agencies have to be justifiably accommodated: not only they may differ, but can be interdependent, complementary, irreconcilable, or simply, independent. We propose different solutions (involving both local and global consistency) to support the adequate merge of the distinct perspectives that inevitably arise during this type of decision making.

Beliefs and Conflicts in a Real World Multi-Agent System

In a real world multiagent system, where the agents are faced with partial, incomplete and intrinsically dynamic knowledge, conflicts are inevitable. Frequently, different agents have goals or beliefs that cannot hold simultaneously. Conflict resolution methodologies have to be adopted to overcome such undesirable occurrences. In this paper we investigate the application of distributed belief revision techniques as the support for conflict resolution in the analysis of the validity of the candidate beams to be produced in the CERN particle accelerators. This CERN multiagent system contains a higher hierarchy agent, the Specialist agent, which makes use of meta-knowledge (on how the con- flicting beliefs have been produced by the other agents) in order to detect which beliefs should be abandoned. Upon solving a conflict, the Specialist instructs the involved agents to revise their beliefs accordingly. Conflicts in the problem domain are mapped into conflicting beliefs of the distributed belief revision system, where they can be handled by proven formal methods. This technique builds on well established concepts and combines them in a new way to solve important problems. We find this approach generally applicable in several domains.

Scenarios Generation for Multi-Agent simulation of Electricity Markets based on Intelligent Data Analysis

This document presents a tool able to automatically gather data provided by real energy markets and to generate scenarios, capture and improve market players’ profiles and strategies by using knowledge discovery processes in databases supported by artificial intelligence techniques, data mining algorithms and machine learning methods. It provides the means for generating scenarios with different dimensions and characteristics, ensuring the representation of real and adapted markets, and their participating entities. The scenarios generator module enhances the MASCEM (Multi-Agent Simulator of Competitive Electricity Markets) simulator, endowing a more effective tool for decision support. The achievements from the implementation of the proposed module enables researchers and electricity markets’ participating entities to analyze data, create real scenarios and make experiments with them. On the other hand, applying knowledge discovery techniques to real data also allows the improvement of MASCEM agents’ profiles and strategies resulting in a better representation of real market players’ behavior. This work aims to improve the comprehension of electricity markets and the interactions among the involved entities through adequate multi-agent simulation.