A challenging task in Chinese collocation extraction is to improve both the precision and recall rate. Most lexical statistical methods including Xtract face the problem of unable to extract collocations with lower fr...
详细信息
Large cluster systems have become widely utilized because they achieve a good performance/cost ratio especially in high performance computing. Although these cluster systems are distributed memory systems, coordinated...
详细信息
Large cluster systems have become widely utilized because they achieve a good performance/cost ratio especially in high performance computing. Although these cluster systems are distributed memory systems, coordinated checkpointing is a promising way to maintain high availability because the computing nodes are tightly connected to one another. However, as the number of computing nodes gets larger, the probability of multi-node failures increases. To tolerate multi-node failures, a large degree of redundancy is required in checkpointing, but this leads to performance degradation. Thus, we propose a new coordinated checkpointing called skewed checkpointing. In this method, checkpointing is skewed every time. Although each checkpointing itself contains only one degree of redundancy, this skewed checkpointing ensures /spl lfloor/log/sub 2/N/spl rfloor/ degrees of redundancy when the number of nodes is N. In this paper, we present the proposed method and an analysis of the performance overhead. Then, this method is applied to a cluster system and compared with other conventional checkpointing schemes. The results reveal the superiority of our method, especially for large cluster systems.
Explicit synchronization mechanisms capable of arbitrary simultaneous barriers are needed to support parallely recursive synchronous MIMD programming, even in step synchronous emulated shared memory machines (ESMM) be...
详细信息
Explicit synchronization mechanisms capable of arbitrary simultaneous barriers are needed to support parallely recursive synchronous MIMD programming, even in step synchronous emulated shared memory machines (ESMM) because control of threads may privately depend on input values. Current synchronization mechanisms fail to support arbitrary simultaneous barriers or are not scalable with future silicon technologies. In this paper, we propose a novel constant execution time barrier synchronization mechanism for scalable ESMMs using active memory. The mechanism is applied to our Eclipse network-on-chip architecture and evaluated briefly.
We survey the models and approaches applied to the middleware services of pervasive computing environments. Architectural styles and patterns introduced in the domain of mobile and wireless computing are also appropri...
详细信息
We survey the models and approaches applied to the middleware services of pervasive computing environments. Architectural styles and patterns introduced in the domain of mobile and wireless computing are also appropriate in pervasive computing. However, the development of pervasive middleware requires more emphasis on interoperability, adaptability and scalability of software services and components. To this aim, we introduce interoperability layers, separation of concerns and ontology-orientation as key enabling technologies to design and maintain agile middleware of pervasive computing environments.
Molecular or deoxyribonucleic acid (DNA) computing is a new research of interest whereby the computation can be done by the use of DNA molecules to encode the computational problem. During the computation, the standar...
详细信息
Molecular or deoxyribonucleic acid (DNA) computing is a new research of interest whereby the computation can be done by the use of DNA molecules to encode the computational problem. During the computation, the standard molecular biology operations are employed and the output that is encoded also by DNA molecules can be printed by electrophoretical fluorescent method. In this paper, a DNA-based computing algorithm for solving unconstraint assignment problem is presented. It is shown that the proposed algorithm uses almost the same laboratory operations as for an instance of Hamiltonian path problem (HPP) of a graph. Since the laboratory experiment for HPP was very promising, the authors believed that ii is possible to verify the proposed DNA-based computing algorithm for unconstraint assignment problem by laboratory experiment.
The paradigm of grid computing has been successfully applied in the domain of computationally-intensive applications supporting scientific research using server-class computers interconnected by wired network. The use...
详细信息
The paradigm of grid computing has been successfully applied in the domain of computationally-intensive applications supporting scientific research using server-class computers interconnected by wired network. The users of the existing and growing number of wireless mobile devices often demand more computational power than their devices can currently provide. We propose using the computational grid paradigm to build a problem-solving environment for wireless mobile devices.
Many are anticipating that grid computing infrastructure, utilities and services will become an integral part of future socio-economical fabric. However, factors such as cost of access, reliability and dependability o...
详细信息
Many are anticipating that grid computing infrastructure, utilities and services will become an integral part of future socio-economical fabric. However, factors such as cost of access, reliability and dependability of the technology undoubtedly influence its adoption and widespread use. IBM are promoting autonomic computing as a solution to alleviate such concerns by delegating software systems' management, security, reliability and performance tuning to the software itself. Based on an on-going research work, focusing on high-assurance concerns of widely distributed systems, this paper outlines a service-oriented approach and associated service infrastructure description languages, which are used to facilitate the construction and management of ad-hoc federated software services.
The continuous improvement of Ethernet technologies is boosting the eagerness of extending their use to cover factory-floor distributed real time applications. Indeed, it is remarkable the considerable amount of resea...
详细信息
The continuous improvement of Ethernet technologies is boosting the eagerness of extending their use to cover factory-floor distributed real time applications. Indeed, it is remarkable the considerable amount of research work that has been devoted to the timing analysis of Ethernet-based technologies in the past few years. It happens, however, that the majority of those works are restricted to the analysis of sub-sets of the overall computing and communication system, thus without addressing timeliness in a holistic fashion. To this end, we address an approach, based on simulation, aiming at extracting temporal properties of commercial-off-the-shelf (COTS) Ethernet-based factory-floor distributed systems. This framework is applied to a specific COTS technology, Ethernet/IP. We reason about the modeling and simulation of Ethernet/IP-based systems, and on the use of statistical analysis techniques to provide useful results on timeliness. The approach is part of a wider framework related to the research project INDEPTH NDustrial-Ethernet ProTocols under Holistic analysis.
Allocating software components while meeting multiple platform resource constraints is crucial for model-based design of large embedded real-time software and automatic design model transformation. In this paper, we p...
详细信息
Allocating software components while meeting multiple platform resource constraints is crucial for model-based design of large embedded real-time software and automatic design model transformation. In this paper, we propose a new method for component allocation using an informed branch-and-bound and forward checking mechanism subject to a combination of resource constraints. We have implemented this method in the automatic integration of reusable embedded software (AIRES) toolkit - which has been developed under the DARPA MoBIES Program - and applied it to an automotive electronic throttle control (ETC) system. Our evaluation based on randomly-generated design models has shown that the proposed method scales well for large and complex embedded real-time software.
Grid computing technologies are adopted to enable wide-spread sharing and coordinated use of network resources. Grid Service is a part of Grid infrastructure enables the integration of service across distributed resou...
详细信息
ISBN:
(纸本)9781595931702
Grid computing technologies are adopted to enable wide-spread sharing and coordinated use of network resources. Grid Service is a part of Grid infrastructure enables the integration of service across distributed resources based on Service-Oriented architecture. In particular, the Open Grid Service Architecture, the new Grid Service model based on Web Service technologies, has been proposed for providing fundamental mechanisms for discovery, monitoring, planning and adapting application behavior. In many kinds of distributed computing, software agents are expected to play an increasing important role for improving performance such as scalability and availability. This paper presents the quality of service (QoS) for multi-agent system, which the concept of agent is applied to Web service technologies. An agent-based hierarchical model is developed to meet the requirement of scalability. The purpose of this model is to improve the performance of the Service Grid system based on Web Service environment by used Multi-process algorithm priority of service.
暂无评论