Demands for efficient and effective utilization of computing resources have derived distributedcomputingsystems of the large scaled such as Grid for scientific computing and Cloud for utility computing. In these dis...
详细信息
ISBN:
(纸本)9783642131189
Demands for efficient and effective utilization of computing resources have derived distributedcomputingsystems of the large scaled such as Grid for scientific computing and Cloud for utility computing. In these distributedcomputingsystems the total capacity of consolidated storages should expand as the amount of user data grows. The user's data are necessarily kept in the distributedcomputing system such as the Grid and the Cloud, which some users don't allow the system to. In this work, an auxiliary storage subsystem is proposed to provide external storages to the distributedcomputing system. The storage subsystem lets the iSCSI devices be connected to the distributedcomputing system as needed. As iSCSI is a recent, standard, and widely deployed protocol for storage networking, the subsystem is able to accommodate many types of storages outside of the distributedcomputing system. The proposed subsystem could transmit data by streams attaining high bit rate of the network capacity in the experiment.
The proceedings contain 57 papers. The topics discussed include: logical topology design with traffic grooming for scheduled demands;the impact of multi-core architectures on task retrieval policies for volunteer comp...
ISBN:
(纸本)9780889867741
The proceedings contain 57 papers. The topics discussed include: logical topology design with traffic grooming for scheduled demands;the impact of multi-core architectures on task retrieval policies for volunteer computing;evaluation of CPU power control and scheduling technique on parallelcomputing environment;clustered assignment of CCQS in information monitoring system supporting conjunctive queries;dense skip graphs as an overlay for P2P networks;multi-connection and multi-core aware all-gather on infiniband clusters;an improved prioritization method for task scheduling in heterogeneous computingsystems;thread pipelining with clustered data communication on commodity CMPS;a distributed algorithm for disjoint paths in star networks;parallel algorithms for a visual text mining platform;and a cache miss analysis for multithreaded architectures.
The UK e-Science EPSRC pilot project Meeting the Design Challenges of nanoCMOS Electronics (***) was funded to address the challenges facing the global electronics semiconductor industry caused by the decreasing size ...
详细信息
ISBN:
(纸本)9780889868205
The UK e-Science EPSRC pilot project Meeting the Design Challenges of nanoCMOS Electronics (***) was funded to address the challenges facing the global electronics semiconductor industry caused by the decreasing size of Complementary Metal Oxide Semiconductor (CMOS) transistors and the atomic variability present in devices manifest at these dimensions. Fundamental problems to be addressed include the modelling, understanding and predicting the effect of differences in the atomic structure of devices on their behaviour, and then using this information to guide electronic circuit and system designers who utilise CMOS components. In this paper we describe the e-Infrastructure that has been developed as part of the nanoCMOS project and outline how it supports large scale high performance computing (HPC) simulations of ensembles of devices which can subsequently be used to model and understand the impact that they have on very large electronic circuits. Key features of this e-Infrastructure include support for very large scale HPC utilization;dealing with federated data sets and associated metadata from multi-level simulations, and addressing challenges related to security and intellectual property protection of data, simulation codes and electronic designs as a whole.
The aim of this work is to assess the performance of a Large Eddy Simulation (LES) in-house code for compressible flows on a 2064 cores Linux computing system at the Inter-University Consortium for the Application of ...
详细信息
ISBN:
(纸本)9780889868205
The aim of this work is to assess the performance of a Large Eddy Simulation (LES) in-house code for compressible flows on a 2064 cores Linux computing system at the Inter-University Consortium for the Application of Super-computing for Universities and Research (CASPUR-Rome-Italy). The code is written in Fortran90 using the Message Passing Interface (MPI) parallelization library. A non-dissipative sixth-order finite difference scheme is employed in order to solve the Favre-filtered governing equations, with a low-pass sixth-order spatial filtering scheme to avoid the growth of high-frequency modes, and a fourth order Runge-Kutta scheme to advance the solution in time. The open Navier-Stokes characteristic boundary conditions are implemented to avoid reflection of pressure waves and the Smagorinsky model is used to compute the subgrid scale turbulent viscosity. The compact-Padè tridiagonal system of equations is solved by employing an efficient parallel partition LU algorithm. In order to assess the performance of the parallelized structure of the code, its speed-up has been evaluated versus the linear speed-up. The model has been validated by performing simulations of an n-heptane non-reacting plane jet issuing into air at high pressure and temperature.
Cauchy-Reed/Solomon is an XOR-based erasure-tolerant coding scheme which is widely used for reliable distributed storage and fault-tolerant memory. A variety of different codes can be specified, depending on the numbe...
详细信息
Intelligent monitoring of large-scale distributed plants is known to exceed capabilities of SCADA systems. In order to perform plant-wide analysis and optimal planning tasks, global self-consistent state of a plant ne...
详细信息
ISBN:
(纸本)9780889868427
Intelligent monitoring of large-scale distributed plants is known to exceed capabilities of SCADA systems. In order to perform plant-wide analysis and optimal planning tasks, global self-consistent state of a plant needs to be constructed at the start of each planning cycle as an input to them. State parameters values deviation shall be verifiably small at the planning horizon. The paper addresses issues associated with automatic computing an informational representation (model) of a global plant state. Its generation is presented as processing of two databases: the information model of a plant, and time series of measured parameters and state-dependant events. The technological transport road map optimization problem for an oil production enterprise is considered as an example. The aspect-oriented programming paradigm is regarded to be the most efficient for implementation of software modules that generate information model of state.
The topologies of Internet and Internet-based information systems have complex network properties. Designing Internet-based virtual computing environment topology with appropriate properties is significant for both th...
详细信息
Many large-scale scientific applications feature distributedcomputing workflows of complex structures that must be executed and transferred in shared wide-area networks consisting of unreliable nodes and links. Mappi...
详细信息
This paper proposes a novel platform for object locating application in the Internet of Things environment. In this platform, objects and inquirers access and query locations using uniform service entry interfaces in ...
详细信息
We present a heterogeneous parallelcomputing system that combines a traditional computer cluster with a broadband network of embedded set-top box (STB) devices. As multiple service operators (MSO) manage millions of ...
详细信息
ISBN:
(纸本)9781450300445
We present a heterogeneous parallelcomputing system that combines a traditional computer cluster with a broadband network of embedded set-top box (STB) devices. As multiple service operators (MSO) manage millions of these devices across wide geographic areas, the computational power of such a massively-distributed embedded system could be harnessed to realize a centrally-managed, energy-efficient parallel processing platform that supports a variety of application domains which are of interest to MSOs, consumers, and the high-performance computing research community. We investigate the feasibility of this idea by building a prototype system that includes a complete head-end cable system with a DOCSIS-2.0 network combined with an interoperable implementation of a subset of Open MPI running on the STB embedded operating system. We evaluate the performance and scalability of our system compared to a traditional cluster by solving approximately various instances of the Multiple Sequence Alignment bioinformatics problem, while the STBs continue simultaneously to operate their primary functions: decode MPEG streams for television display and run an interactive user interface. Based on our experimental results and given the technology trends in embedded computing we argue that our approach to leverage a broadband network of embedded devices in a heterogeneous distributed system offers the benefits of both parallelcomputing clusters and distributed Internet computing.
暂无评论