This paper describes a middleware, called ABEcity, for agent-based autonomous processing. It is designed to automatically generate heterogeneous agents and to support them in executing parallel and heterogeneous tasks...
详细信息
ISBN:
(纸本)1892512459
This paper describes a middleware, called ABEcity, for agent-based autonomous processing. It is designed to automatically generate heterogeneous agents and to support them in executing parallel and heterogeneous tasks in distributed environments. We depict its architecture, components and show how agents in ABEcity represent and handle distributed and parallelprocessing. The hp-Agents, naming service, communication, migration, interaction, resource and workload control, and security are described.
Improvements in the processing speed of multiprocessors are outpacing improvements in the speed of disk hardware. parallel disk I/O subsystems have been proposed as one way to close the gap between processor and disk ...
详细信息
Improvements in the processing speed of multiprocessors are outpacing improvements in the speed of disk hardware. parallel disk I/O subsystems have been proposed as one way to close the gap between processor and disk speeds. In a previous paper we showed that prefetching and caching have the potential to deliver the performance benefits of parallel file systems to parallelapplications. In this paper we describe experiments with practical prefetching policies that base decisions only on on-line reference history, and that can be implemented efficiently. We also test the ability of those policies across a range of architectural parameters.
Mutual exclusion is a fundamental problem in concurrent, parallel, and distributedprocessing systems. In this paper we present an efficient algorithm to solve mutual exclusion problem. The algorithm is a modification...
详细信息
ISBN:
(纸本)1932415262
Mutual exclusion is a fundamental problem in concurrent, parallel, and distributedprocessing systems. In this paper we present an efficient algorithm to solve mutual exclusion problem. The algorithm is a modification of a popular algorithm. Our modifications are simple and hence preserves the simplicity of the original algorithm almost intact.
In this paper, we examine the suitability of CORBA-based solutions for meeting application requirements in the field of distributedparallel programming. We outline concepts defined within CORBA which are helpful for ...
详细信息
ISBN:
(纸本)0769505686
In this paper, we examine the suitability of CORBA-based solutions for meeting application requirements in the field of distributedparallel programming. We outline concepts defined within CORBA which are helpful for the development of parallelapplications, and we describe which programming techniques are at hand for this purpose. Subsequently, we present an Object Group Service which facilitates the development of CORBA-based, distributed and parallel software applications. Moreover, we introduce some basic ideas of how the Unified modeling Language (UML) can be used for modeling parallelapplications.
In this modern era of knowledge explosion, one of the most exciting research areas is Artificial Intelligence (AI). In this paper we investigate more specifically the theory and design of a new area of AI labeled dist...
详细信息
ISBN:
(纸本)1892512459
In this modern era of knowledge explosion, one of the most exciting research areas is Artificial Intelligence (AI). In this paper we investigate more specifically the theory and design of a new area of AI labeled distributed intelligent agents, whose development encompasses a broad variety of applications. From robots to lunar landers, this technology holds great promise for providing intelligence to systems, that has either required human interaction or whose programming was so complex and expensive as to render it useless. The design specifications given in this paper will provide a research and development framework for a marriage of distributed artificial intelligence, intelligent agents, and distributedparallelprocessing called distributed Intelligent Agents (DIA).
The purpose of the ArmCluster Project is creation of a distributed information-processing network in the Republic of Armenia, in order to provide the scientists and engineers in the South Caucasus region with high-per...
详细信息
ISBN:
(纸本)1932415262
The purpose of the ArmCluster Project is creation of a distributed information-processing network in the Republic of Armenia, in order to provide the scientists and engineers in the South Caucasus region with high-performance information sources and databases. A brief introduction of the project is given.
This paper reports the development of the Concerto platform, which is dedicated to supporting the deployment of resource-aware parallel Java components on heterogeneous distributed platforms, such as pools of workstat...
详细信息
ISBN:
(纸本)1932415262
This paper reports the development of the Concerto platform, which is dedicated to supporting the deployment of resource-aware parallel Java components on heterogeneous distributed platforms, such as pools of workstations in labs or offices. We propose a basic model of a parallel Java component and present some tools that facilitate the management and the deployment of such a component on a distributed architecture. The Concerto platform also defines and implements an open and extensible framework dedicated to the observation of the distributed environment on which a component is deployed Support is provided for modeling the execution environment as a set of resources and for monitoring these resources in a distributed way.
The programming model of the current distributed systems such as middleware CORBA or JAVA RMI is based on remote procedure call. Without the support of high-level API for coordination, developing real-world distribute...
详细信息
ISBN:
(纸本)1932415262
The programming model of the current distributed systems such as middleware CORBA or JAVA RMI is based on remote procedure call. Without the support of high-level API for coordination, developing real-world distributedapplications in this model is difficult. In this paper we present an algorithm for multiparty interaction, a key abstraction of distributed coordination in the future distributed programming models, and its JAVA implmentation called MULTIPI.
parallel and distributed programs exhibit nondeterministic behavior, whenever two or more processes are competing for a race condition, and the outcome of this race determines the subsequent execution of the program. ...
详细信息
ISBN:
(纸本)1932415262
parallel and distributed programs exhibit nondeterministic behavior, whenever two or more processes are competing for a race condition, and the outcome of this race determines the subsequent execution of the program. While race conditions themselves are not errors per se, their handling is tricky, making program analysis difficult if not impossible. This paper describes the effects of non-determinism as experienced with a real-world parallel program from theoretical physics, observations and remarks about race conditions in parallel and distributed programs in general, and a corresponding solution as implemented in the MAD environment for the message passing interface standard MPI.
暂无评论