To achieve high performance on modern computers, it is vital to map algorithmic parallelism to that inherent in the hardware. From an application developer's perspective, it is also important that code can be main...
详细信息
ISBN:
(纸本)9781479961238
To achieve high performance on modern computers, it is vital to map algorithmic parallelism to that inherent in the hardware. From an application developer's perspective, it is also important that code can be maintained in a portable manner across a range of hardware. Here we present targetDP (target Data Parallel), a lightweight programming layer that allows the abstraction of data parallelism for applications that employ structured grids. A single source code may be used to target boththread level parallelism (TLP) and instruction level parallelism (ILP) on either SIMD multi-core CPUs or GPU-accelerated platforms. targetDP is implemented via standard C preprocessor macros and library functions, can be added to existing applications incrementally, and can be combined with higher-level paradigms such as MPI. We present CPU and GPU performance results for a benchmark taken from the lattice Boltzmann application that motivated this work. these demonstrate not only performance portability, but also the optimisation resulting from the intelligent exposure of ILP.
thisvolumecontainstheinvitedandregularpaperspresentedat TCS 2010,the 6thIFIP International conference on theoretical Computer Science, organised by IFIP Tech- cal Committee 1 (Foundations of Computer Science) and IFIP...
ISBN:
(纸本)9783642423116
thisvolumecontainstheinvitedandregularpaperspresentedat TCS 2010,the 6thIFIP International conference on theoretical Computer Science, organised by IFIP Tech- cal Committee 1 (Foundations of Computer Science) and IFIP WG 2.2 (Formal - scriptions of programming Concepts) in association with SIGACT and EATCS. TCS 2010 was part of the World Computer Congress held in Brisbane, Australia, during September 20-23, 2010 ( ). TCS 2010 is composed of two main areas: (A) Algorithms, Complexity and Models of Computation, and (B) logic, Semantics, Speci?cation and Veri?cation. the selection process led to the acceptance of 23 papers out of 39 submissions, *** Committee discussion was held electronically using Easychair. the invited speakers at TCS 2010 are: Rob van Glabbeek (NICTA, Australia) Bart Jacobs (Nijmegen, the Netherlands) Catuscia Palamidessi (INRIA and LIX, Paris, France) Sabina Rossi (Venice, Italy) James Harland (Australia) and Barry Jay (Australia) acted as TCS 2010 Chairs. We take this occasion to thank the members of the Programme Committees and the external reviewers for the professional and timely work; the conference Chairs for their support; the invited speakers for their scholarly contribution; and of course the authors for submitting their work to TCS 2010.
Teaching new technologies has two facets. A first facet is the pleasure to discover new concepts with beginners. the second facet is the difficulty to create the right training for students who will apply this knowled...
详细信息
ISBN:
(纸本)9781612083285
Teaching new technologies has two facets. A first facet is the pleasure to discover new concepts with beginners. the second facet is the difficulty to create the right training for students who will apply this knowledge through an internship by the end of the training. Since January 2013, our challenge has been to prepare a new training about nomadic device and Android programming. Our training is structured as a master's degree in our UPEC University (Paris). this document presents both aspects of this master degree: the construction of the training and the perception of teaching activities by the students. We observe that the use of a tablet or a smart phone during the lessons is not enough. Pragmatic approach is limited even if it is a help for the participation of students, and a motivation for teachers and students in a computer science context where it is crucial to explore new domains. Learning with device provides a more practical teaching. But its use has to be completed by knowledge validation with it. Because devices take place into teaching approach, the whole process of teaching has to evolve into a more interactive exchange between teachers and students.
We present the architecture of the Motion Data Editor (MDE) real-time development framework for multi-modal motion data management, visualization and analysis. there is an emerging need for such tools due to the capab...
详细信息
ISBN:
(纸本)9783319054582;9783319054575
We present the architecture of the Motion Data Editor (MDE) real-time development framework for multi-modal motion data management, visualization and analysis. there is an emerging need for such tools due to the capability of recording and computing large number of motion modalities with high precision, synchronized in time domain. Such tools should be efficient, easy to use and flexible to apply to various data types and algorithms. Proposed Motion Data Editor (MDE) is dedicated to general data processing, and supports most of the common functionalities during data analysis. We discuss the most important functional requirements and present selected elements of system architecture: core data types, functionality and processing logic elements. the MDE with Human Motion Laboratory (HML) and cloud based Human Motion Database (BDR) [3] constitute collaborative environment for acquisition and analysis of multi-modal synchronized motion data for medical research and entertainment.
the paper begins with discussion on Aspect-Oriented programming (AOP), including its concept, thought, development, and advantage in handling crosscutting concerns, followed by description of the theory of constraints...
详细信息
ISBN:
(纸本)9781479977659
the paper begins with discussion on Aspect-Oriented programming (AOP), including its concept, thought, development, and advantage in handling crosscutting concerns, followed by description of the theory of constraints. Furthermore, combined AOP with first-order logic language, a schema on detection of the business constraints in requirement analysis is presented, which is simple and practicable, thus making optimal designs for improving software robustness.
the proceedings contain 59 papers. the special focus in this conference is on Engineering Virtualized Systems, Statistical Model Checking, Risk-Based Testing, Medical Cyber-Physical Systems, Scientific Workflows, Eval...
ISBN:
(纸本)9783662452301
the proceedings contain 59 papers. the special focus in this conference is on Engineering Virtualized Systems, Statistical Model Checking, Risk-Based Testing, Medical Cyber-Physical Systems, Scientific Workflows, Evaluation and Reproducibility of Program Analysis and Industrial Track. the topics include: Introduction to track on engineering virtualized services;fault model design space for cooperative concurrency;towards the typing of resource deployment;fully abstract operation contracts;an effective heuristic for adaptive importance splitting in statistical model checking;a formalism for stochastic adaptive systems;a review of statistical model checking pitfalls on real-time stochastic models;using statistical model checking for measuring systems;blocking advertisements on android devices using monitoring techniques;a risk assessment framework for software testing;combining risk analysis and security testing;risk-based vulnerability testing using security test patterns;challenges for the dynamic interconnection of medical devices;temporal logic based monitoring of assisted ventilation in intensive care patients;meta-analysis of disjoint sets of attributes in large cohort studies;a visual programming approach to beat-driven humanoid robot dancing;automatic annotation of bioinformatics workflows with biomedical ontologies;modeling and verifying communication systems;simple management of high assurance data in long-lived interdisciplinary healthcare research;domain-specific business modeling withthe business model developer;semantic heterogeneity in the formal development of complex systems;modelling and verifying an evolving distributed control system using an event-based approach and requirements driven data warehouse design.
the objective of this paper is to propose an intelligent robust controller to control the sun seeker system which is mounted on a space vehicle. the control objective is to maximize the amount of current produced by t...
详细信息
the objective of this paper is to propose an intelligent robust controller to control the sun seeker system which is mounted on a space vehicle. the control objective is to maximize the amount of current produced by the photo-voltaic cells which are mounted on the sun seeker system. the PV cells produce electric current proportional to the amount of the sun light incident over them. thus the aim is to control the position of the sun seeker system such that the position of the PV cells enables them to produce maximum current. the result of the proposed controller is then compared for robustness withthe classical proportional integral derivative (PID) controller and the modern pole placement technique (PPT) based controller. the simulated results using MATLAB clearly indicate that the proposed controller is more robust as compared to the other controllers.
the diagnosis of incipient fault is very important for power transformer condition monitoring. the incipient faults are monitored by conventional and artificial intelligence (AI) based models. In this paper, the GEP h...
详细信息
ISBN:
(纸本)9781479960439
the diagnosis of incipient fault is very important for power transformer condition monitoring. the incipient faults are monitored by conventional and artificial intelligence (AI) based models. In this paper, the GEP has been utilized to identify the incipient faults in an oil-immersed power transformer. Its performance is compared with traditional IEC/IEEE and AI methods (i.e. ANN and fuzzy logic). the juxtaposition of fault classification of ANN and FL method notify that proposed approach is much swiftly. the desired test analysis of experimental data from working transformers in the Northern Power Grid of India has been executed to present the robustness of evaluated incipient faults for wide changes in operational and loading conditions perturbations.
In this report, we propose to give a review of the most used clustering methods in the literature. First, we give an introduction about clustering methods, how they work and their main challenges. Second, we present t...
详细信息
In this report, we propose to give a review of the most used clustering methods in the literature. First, we give an introduction about clustering methods, how they work and their main challenges. Second, we present the clustering methods with some comparisons including mainly the classical partitioning clustering methods like well-known k-means algorithms, Gaussian Mixture Models and their variants, the classical hierarchical clustering methods like the agglomerative algorithm, the fuzzy clustering methods and Big data clustering methods. We present some examples of clustering algorithms comparison. Finally, we present our ideas to build a scalable and noise insensitive clustering system based on fuzzy type-2 clustering methods.
the proceedings contain 12 papers. the topics discussed include: properties of answer set programming with convex generalized atoms;hybrid automated reasoning tools: from black-box to clear-box integration;aspartame: ...
the proceedings contain 12 papers. the topics discussed include: properties of answer set programming with convex generalized atoms;hybrid automated reasoning tools: from black-box to clear-box integration;aspartame: solving constraint satisfaction problems with answer set programming;a functional view of strong negation in answer set programming;an algebra of causal chains;query answering in object oriented knowledge bases in logicprogramming: description and challenge for ASP;the DIAMOND system for argumentation: preliminary report;a system for interactive query answering with answer set programming;and generating shortest synchronizing sequences using answer set programming.
暂无评论