Web services are generally intended to be accessed by an unlimited number of users. The service provider decides when and how they should be modified or extended. But this unidirectional approach may not satisfy the n...
详细信息
Web services are generally intended to be accessed by an unlimited number of users. The service provider decides when and how they should be modified or extended. But this unidirectional approach may not satisfy the needs of all the users. In fact, such a dependency forces users to create their own adaptations and proxy services. This situation cause, on one hand, the need to develop private versions of the service which could be otherwise public and therefore reusable. On the other hand, the use of proxies introduce some disadvantages, namely the additional overhead and network load cost, and the impossibility to cope with end-to-end adaptations. To overcome these problems we propose a community model for web services, where users can deploy adaptations in the hosting service without disturbing other users. In this paper we present the framework for this model and the experimental results of the developed prototype. These results show that the adaptations cause an imperceptible performance overhead.
computer networks and large-scale systems are growing and becoming increasingly hard to manage, specially in the software front, but also in hardware, and their management information more complicated and hard to inte...
详细信息
computer networks and large-scale systems are growing and becoming increasingly hard to manage, specially in the software front, but also in hardware, and their management information more complicated and hard to integrate in a seamless manner. This paper presents a framework that a) allows the extraction of configuration from its original format, b) models it (including relations and inheritance) via the CIM object-oriented information model, and c) persists and retrieves configuration transparently, independently of the repository, and retaining schema constraints. These steps are facilitated by standard, declarative XSLT templates that allow fast and natural adaptation to new configuration domains, allowing the integration and representation of all configuration information.
The aim of this work is to provide a high performance air quality simulation using the sulphur transport Eulerian model 2 (STEM-II) program. First of all we optimize the sequential program with the aim of increasing d...
详细信息
The aim of this work is to provide a high performance air quality simulation using the sulphur transport Eulerian model 2 (STEM-II) program. First of all we optimize the sequential program with the aim of increasing data locality. Then, the optimized program is parallelized using OpenMP shared-memory directives. Experimental results on a 32-processor SGI Origin 2000 show that the parallel program achieves important reductions in the execution times.
The CrossGrid project develops, implements and exploits new Grid components for interactive compute and data intensive applications like simulation and visualization for surgical procedures, flooding crisis team decis...
详细信息
A loop with irregular assignment computations contains loop-carried output data dependences that can only be detected at run-time. In this paper, a load-balanced method based on the inspector-executor model is propose...
详细信息
A loop with irregular assignment computations contains loop-carried output data dependences that can only be detected at run-time. In this paper, a load-balanced method based on the inspector-executor model is propose...
ISBN:
(纸本)9783540241287
A loop with irregular assignment computations contains loop-carried output data dependences that can only be detected at run-time. In this paper, a load-balanced method based on the inspector-executor model is proposed to parallelize this loop pattern. The basic idea lies in splitting the iteration space of the sequential loop into sets of conflict-free iterations that can be executed concurrently on different processors. As will be demonstrated, this method outperforms existing techniques. Irregular access patterns with different load-balancing and reusability properties are considered in the experiments.
The use of Java for parallel programming on clusters according to the message-passing paradigm is an attractive choice. In this case, the overall application performance will largely depend on the performance of the u...
详细信息
A decoupled front-end architecture relies on a separate fetch predictor that is critical in determining the fetch bandwidth. The Fetch Target Buffer (FTB) is a multilevel fetch predictor design. In this paper we propo...
详细信息
This paper summarizes the content and application of the SEED2002 project. SEED2002 is a university/industry collaboration to enhance and reform the education in the field of semi-custom/full-custom ASIC design at the...
详细信息
The use of Java for parallel programming on clusters according to the message-passing paradigm is an attractive choice. In this case, the overall application performance will largely depend on the performance of the u...
详细信息
暂无评论