The increasingly widespread availability of multi-processor computers has resulted in interest to design software that can make a full use of them. This is specially important in research andengineering applications,...
详细信息
ISBN:
(纸本)9781905088423
The increasingly widespread availability of multi-processor computers has resulted in interest to design software that can make a full use of them. This is specially important in research andengineering applications, which typically require significant computing power. In this paper we present a parallel finite element solver for the Navier-Stokes equations, based on algebraic subgrid scale stabilization and using a generalized Newmark scheme for time iteration. This solver is implemented using two different parallel strategies, OpenMP and MPI, and used to solve a practical example. We are interested in the performance of each parallel implementation, which we will test by solving a practical example.
In this paper, we study the benefits and overheads of lazy MapReduce processing, where the input data is partitioned and only the smallest subset of these partitions are processed to meet a user's need at any time...
详细信息
The proceedings contain 8 papers. The topics discussed include: the case for being lazy: how to leverage lazy evaluation in MapReduce;cloud resource usage - extreme distributions invalidating traditional capacity plan...
ISBN:
(纸本)9781450306997
The proceedings contain 8 papers. The topics discussed include: the case for being lazy: how to leverage lazy evaluation in MapReduce;cloud resource usage - extreme distributions invalidating traditional capacity planning models;experiences using cloud computing for a scientific workflow application;adaptive rate stream processing for smart grid applications on clouds;an automated approach to cloud storage service selection;Neptune: a domain specific language for deploying HPC software on cloud platforms;Magellan: experiences from a science cloud;and Cumulus: an open source storage cloud for science.
This paper introduces a new approach, based on state machines, for distributed frameworks, that is able to support both distributed simulation and computational steering. Existing frameworks are in general dedicated t...
详细信息
ISBN:
(纸本)9781618040343
This paper introduces a new approach, based on state machines, for distributed frameworks, that is able to support both distributed simulation and computational steering. Existing frameworks are in general dedicated to certain projects or they only take care of a part of the issues related to this field. Our approach is scalable, flexible, supports load balancing and tasks migration while providing a key feature for any complex system: safety.
To date, software development for wireless sensor network nodes is still characterized by low-level ad-hoc programming to a large extent. This short-paper argues for a methodologically more systematic approach on the ...
详细信息
The proceedings contain 59 papers. The topics discussed include: proprietary code to non-proprietary benchmarks: synthesis techniques for scalable benchmarks;performance analysis of domain specific visual models;perfo...
ISBN:
(纸本)9781450305198
The proceedings contain 59 papers. The topics discussed include: proprietary code to non-proprietary benchmarks: synthesis techniques for scalable benchmarks;performance analysis of domain specific visual models;performance modeling in MapReduce environments: challenges and opportunities;computing first passage time distributions in stochastic well-formed nets;detection and solution of software performance antipatterns in Palladio architectural models;an approach for scalability-bottleneck solution: identification and elimination of scalability bottlenecks in a DBMS;experience building non-functional requirement models of a complex industrial architecture;relative roles of instruction count and cycles per instruction in WCET estimation;an automatic trace based performance evaluation model building for paralleldistributedsystems;and real-world performance modeling of enterprise service oriented architectures: delivering business value with complexity and constraints.
OOSol is a software project whose main aim is the development of state-of-the-art library of subroutines that will facilitate development and experiments with numerical methods and tools for computer-aided engineering...
详细信息
ISBN:
(纸本)9781905088423
OOSol is a software project whose main aim is the development of state-of-the-art library of subroutines that will facilitate development and experiments with numerical methods and tools for computer-aided engineering problems. The title OOSol was created as an abbreviation of words object oriented solvers. This indicates that the library will be based on the object oriented technology. The goal is to involve the object oriented technology not only in implementation of data structures as usual but also to exploit main features of this technology for implementation of numerical algorithms. In the last months, the modules for parallel anddistributed computing have been developed. In this paper, we firstly present overview to OOSol library and its data structures, with emphasis to distributed programming. Secondly, we present the parallel scalability tests on given numerical example of two membranes using standard conjugate gradient algorithm, using the OOSol library.
The rapid advancement in the collection of environmental data, geospatial analysis and modeling software calls for new GIS-enabled cyberinfrastructure (CI) capable of support large scale integrated data and modeling a...
详细信息
作者：
Kraines, StevenFuture Center Initiative
Environmental Science Research Building University of Tokyo Rm 277 5-1-5 Kashiwa-no-ha Kashiwa-shi 277-8563 Chiba-ken Japan
In order to conduct simulation studies of highly complex problems, such as integrated environmental assessment of technologies, policies and other measures for making urban ecosystems more sustainable, integration of ...
详细信息
ISBN:
(纸本)9789898425829
In order to conduct simulation studies of highly complex problems, such as integrated environmental assessment of technologies, policies and other measures for making urban ecosystems more sustainable, integration of knowledge in a dynamic form from a wide range of knowledge domains is essential. We have proposed that computational models are appropriate representations of expert knowledge for such integration. Building on previously introduced concepts andsoftware prototypes, we have been designing a dynamic computational modeling platform for studying the integrated effects of supply side and demand side technologies and policies to reduce environmental impacts and consumption of resources caused by activities in urban ecosystems such as Tokyo, Japan. This platform has been used to evaluate scenarios that include the introduction of roof-top photovoltaics, a solid oxide fuel cell combined with a gas turbine topping cycle (SOFC/GT), energy conservation measures in the residential and commercial building sectors, and waste processing and recycling systems. Here we describe the softwareengineering issues associated with the construction of this model integration platform and provide examples of the techniques we have developed to address those issues.
The proceedings contain 15 papers. The topics discussed include: an internationally distributed cloud for science: the cloud-enabled space weather platform;modeling cloud failure data: a case study of the virtual comp...
ISBN:
(纸本)9781450305822
The proceedings contain 15 papers. The topics discussed include: an internationally distributed cloud for science: the cloud-enabled space weather platform;modeling cloud failure data: a case study of the virtual computing lab;Uni4Cloud: an approach based on open standards for deployment and management of multi-cloud applications;application migration to cloud: a taxonomy of critical factors;cloud adoption: a goal-oriented requirements engineering approach;evaluating cloud computing in the NASA DESDynI ground data system;a cloud-enabled regional climate model evaluation system;a tale of migration to cloud computing for sharing experiences and observations;a MapReduce workflow system for architecting scientific data intensive applications;and an application architecture to facilitate multi-site clinical trial collaboration in the cloud.
暂无评论