The convergence of the two widely used parallel programming paradigms, shared- memory and distributed-shared-memory parallel programming models, into a unified parallel programming model is crucial for parallel comput...
详细信息
The convergence of the two widely used parallel programming paradigms, shared- memory and distributed-shared-memory parallel programming models, into a unified parallel programming model is crucial for parallel computing to become the next mainstream programming paradigm. We study the design differences and the performance issues of two parallel programming models: a shared-memory programming model (OpenMP) and a distributed-shared programming model (BSP). The study was carried out by designing a compiler for translating BSP parallel programs to an OpenMP programming model called BSP2OMP. Analysis of the compiler outcome, and of the performance of the compiled programs, show that the two models are based on very similar underlying principles and mechanisms.
As today's computing trends are moving towards the cloud, meeting the increasing computational demand while minimizing the energy costs in data centers has become essential. This work introduces two adaptive techn...
详细信息
Model based development within the area of hardware-software embedded systems co-design has been used as a major attitude to support embedded system complexity increasing during the last decade. Several modeling forma...
详细信息
Model based development within the area of hardware-software embedded systems co-design has been used as a major attitude to support embedded system complexity increasing during the last decade. Several modeling formalisms are very well matched for usage within this area. One of the main goals of the work presented in this paper is to contribute for the usage of Petri nets as the system-level specification language within model-based development of embedded systems using co-design techniques. System model partitioning into modules and components ready to be implemented in hardware or software and be deployed into a distributed platform is a major concern within co-design methodologies. A net operation, called net splitting, is used in this paper in order to decompose a Petri net model into several Petri net submodels, which will be executed in parallel using synchronous communication channels. Those Petri net sub-models are seen as modules or components ready to be used for distributed execution of the initial model. In this paper, we propose a set of techniques to allow module replication and their composition in new situations, supporting module reutilization, which has a key impact factor within engineering activities. The proposed techniques rely on keeping the module interface unchanged in order to support the integration of the module in new situations.
Web Services are becoming more and more fundamental building blocks of Web-based distributed applications and a core technology for grid systems. Due to their flexibility, Web Services easily combine, in a common and ...
详细信息
Web Services are becoming more and more fundamental building blocks of Web-based distributed applications and a core technology for grid systems. Due to their flexibility, Web Services easily combine, in a common and coherent framework, ubiquitous computing with heterogeneous applications composed of different kinds of resources and, typically distributed in many organizations. We expect that this technology will follow the same evolution paths that have characterized other technologies so far, with some specificity due to the openness and size of the application context. In this connection, optimizations tied to invocations and workflows are assuming a primary role in Web Services research. The synchronous request/reply nature of the most diffused underling protocol (HTTP) introduces several restrictions in many application scenarios. On the other hand, asynchronous interactions are allowed by using message oriented middleware platforms, like JMS, which are typically harder to handle than object- and process-oriented middleware. In this paper, we propose a first implementation of a module that allows for dynamic Web Services invocations, which, on the basis of metadata added to WSDL, is able to select the most appropriate invocation technique for calling a Web Services operation.
Many applications in comparative genomics lend themselves to implementations that take advantage of common high-performance features in modern microprocessors. However, the common suggestion that a data-parallel, mult...
详细信息
Many applications in comparative genomics lend themselves to implementations that take advantage of common high-performance features in modern microprocessors. However, the common suggestion that a data-parallel, multithreaded, or high-throughput implementation is possible often ignores the complexity of actually creating such software. In this paper, we present a data-parallel algorithm for a classic comparative genomics algorithm, the dot plot, along with a multiprocessor extension. For large genomic comparisons, these new algorithms achieve speedups of up to 14.4x over the sequential version. This speedup introduces the opportunity of performing full pairwise comparisons on entire genomes on a much larger scale than previously possible. We also present the experimental, model-driven approach used to develop the algorithm that allowed us to carefully study and evaluate implementation options and fully understand the parameters affecting its performance.
The proceedings contain 208 papers. The topics discussed include: design and evaluation of knowledge-based tool for course exemption in institution of higher learning;social software enhanced e-learning system to supp...
ISBN:
(纸本)9781467319386
The proceedings contain 208 papers. The topics discussed include: design and evaluation of knowledge-based tool for course exemption in institution of higher learning;social software enhanced e-learning system to support knowledge sharing among students: Malaysian higher education institutions perspectives;theoretical framework for multi-agent collaborative knowledge sharing for competitiveness of institutions of higher learning (IHL) in Malaysia;design of a medication reminder and feedback system for Thai elders;a framework of a childhood obesity intervention using persuasive web-mobile technology;a deadlock-free dimension order routing for hierarchical 3D-mesh network;an approach towards intrusion detection using PCA feature subsets and SVM;parallel heuristic graph matching algorithm for task assignment problem in distributed computing systems;and in support of hydrocarbon exploration and discovery using clouds.
Nowadays parallelsoftware system is very common and practical. However, it is difficult to test parallelsoftware, because the state space of parallelsoftware is very large. Therefore, a parallel model simplificatio...
详细信息
ISBN:
(纸本)9781538637913
Nowadays parallelsoftware system is very common and practical. However, it is difficult to test parallelsoftware, because the state space of parallelsoftware is very large. Therefore, a parallel model simplification method based on CPN (Color Petri Net) is proposed. Based on the original CPN, the CPN model for the tested behavior(Tested Behavior of CPN, TBoCPN) is proposed. The target of the test is described as the tested behavior. The relevant behavior is described as the behavior related to the tested behavior, then, the homogeneous concurrent branch group and the selection branch set are divided. Finally, the branches of the concurrent branch group and the selected branch set, which satisfied the condition of algorithm, are sequentially processed by the inhibitor arcs. The experiment shows that the reduction rate is at least 60%, and before and after the reduction, the full coverage test path generated by the tested behavior is not affected, thus proving that the method is an effective test method.
A new class of Java multithreading based parallel approximate inverse preconditioning is introduced, for efficiently solving sparse arrow-type linear systems. The parallel Explicit Preconditioned Biconjugate Conjugate...
详细信息
ISBN:
(纸本)9780769529172;0769529178
A new class of Java multithreading based parallel approximate inverse preconditioning is introduced, for efficiently solving sparse arrow-type linear systems. The parallel Explicit Preconditioned Biconjugate Conjugate Gradient - STAB method for shared memory systems is presented in order to examine the parallel behavior of this scheme using explicit approximate inverses as the suitable preconditioner. Design and implementation issues of Java's multithreading techniques are also discussed. The performance in terms of speedups and parallel efficiencies of the method, using Java multithreading, is illustrated by solving sparse arrow-type linear systems. Static and dynamic workload scheduling systems implemented in Java and the results of their use are presented and discussed.
Quality of service in IP networks necessitates the use of performance management. As Internet continues to grow exponentially, a management system should be scalable in terms of network size, speed and number of custo...
详细信息
ISBN:
(纸本)0769521355
Quality of service in IP networks necessitates the use of performance management. As Internet continues to grow exponentially, a management system should be scalable in terms of network size, speed and number of customers subscribed to value-added services. This article proposes a flexible, scalable, self-adapting model for managing large-scale distributed network. In this model, Web services framework is used to build the software architecture and XML is used to build the data exchange interface. Policy-based hierarchical event-processing mechanism presented by this paper can efficiently balance the loads and improve the flexibility of the system. The prediction algorithm adopted by this model can predict the network performance more effectively and accurately.
A back-to-back HVDC converter system is typically used to connect two separate electrical grids or a remote large-scale renewable energy system with the electrical grid through a DC bus. The DC bus typically contains ...
详细信息
暂无评论