We present the Lightweight Information Validation Environment, LIVE as asolution to the high complexity and data sizes of modern day computational science applications. LIVE is a data workspace that facilitates the cr...
详细信息
ISBN:
(纸本)1595936734
We present the Lightweight Information Validation Environment, LIVE as asolution to the high complexity and data sizes of modern day computational science applications. LIVE is a data workspace that facilitates the creation of dynamic data processing overlays we call I/O graphs. We use LIVE as aplatform for dynamic extension of scientific applications using lightweight data extraction, runtime discovery and flexible data selection.
Traditional network design incorporates a failure-recovery model in order to allow calculation of problems independent of knowledge of the network tool layer this paper explores the possibilities of improving the calc...
详细信息
ISBN:
(纸本)0769516262
Traditional network design incorporates a failure-recovery model in order to allow calculation of problems independent of knowledge of the network tool layer this paper explores the possibilities of improving the calculation throughput by constructing a tool for the specific solution of problems which have an inherent ability to deal with partial calculation failure. Using a modified Genetic Algorithm as the client tool, the amount of information the network layer needs to have is brought to an extremely minimal level;this allows for a large scalability factor of the tool due to the reduction of network management tables.
As one of possibilities of high-performancecomputing (HPC) technologies for electromagnetic microwave simulation, the authors have been working in development of finite-difference time-domain/finite-integration techn...
详细信息
As one of possibilities of high-performancecomputing (HPC) technologies for electromagnetic microwave simulation, the authors have been working in development of finite-difference time-domain/finite-integration technique (FDTD/FI)-dedicated computers. there are two types of FDTD/FIT-dedicated computerarchitectures, extremely high-performance-oriented and large-scale simulation-oriented machines. this paper presents an improved architecture of the large-scale simulation-oriented machine for higher performance computation.
Particle-based models are widespread in the field of computer graphics and are mostly used in soft-body dynamics, for simulating surfaces such as cloth, fluids and biologic tissue. As model resolution and scenario com...
详细信息
ISBN:
(纸本)9781479984480
Particle-based models are widespread in the field of computer graphics and are mostly used in soft-body dynamics, for simulating surfaces such as cloth, fluids and biologic tissue. As model resolution and scenario complexity increases, the computation required for these particular applications becomes overwhelming for a single processing unit, especially when interactivity is required, thus parallelization must be employed in order to provide a fast, flexible and scalable simulation environment. high-performancecomputingarchitectures such as graphics clusters may provide the parallel computing and rendering power required, but the distributed and remote nature of the computation and rendering process introduce specific challenges that must be tackled. We propose a parallel, distributed, modular system architecture for a particle-based simulator on GPU clusters, encapsulating powerful parallel and distributed processing, distributed rendering and remote interaction techniques, for flexible, fast simulation of large models and complex scenarios. For validating and evaluating the proposed architecture, we perform a visual comparison of two largely used numeric integration methods, namely the explicit Velocity Verlet and implicit Euler integration techniques.
Recent developments in the international arena has meant the technology is now mature enough to bring together those required for the implementation of a grid computing facility. this paper examines the requirements a...
详细信息
ISBN:
(纸本)0769517722
Recent developments in the international arena has meant the technology is now mature enough to bring together those required for the implementation of a grid computing facility. this paper examines the requirements and applications for an eScience infrastructure with particular reference to developments in Europe.
Beowulf clusters, on face value, offer the potential of a viable cost effective alternative for the provision of highperformancecomputing. In this paper we compare the performance of Beowulf clusters built from comm...
详细信息
ISBN:
(纸本)0769516262
Beowulf clusters, on face value, offer the potential of a viable cost effective alternative for the provision of highperformancecomputing. In this paper we compare the performance of Beowulf clusters built from commodity "off the shelf" components in the support of major research and production codes, with current high-end hardware such as the IBM SP, Compaq AlphaServer SC and SGI Origin 3800. the results concentrate on the application area of computational chemistry. Benchmark data on six commodity-based systems (CS1-CS6) featuring Intel, AMD Athlon and Alpha CPU architectures coupled to traditional Beowulf interconnect, such as Myrinet and Ethernet, are presented. Furthermore, we provide performance data on systems utilising the Quadrics QSNet interconnect technology, and initial results from a prototype of the Cray Supercluster.
Tightly coupled parallel applications are increasingly run in Grid environments. Unfortunately, on many Grid sites the ability of machines to create or accept network connections is severely limited by ?rewalls, netwo...
详细信息
ISBN:
(纸本)1595936734
Tightly coupled parallel applications are increasingly run in Grid environments. Unfortunately, on many Grid sites the ability of machines to create or accept network connections is severely limited by ?rewalls, network address translation (NAT)or non-routed networks. Multi homing further complicates connection setup and machine identi?cation. Although ad-hoc solutions exist for some of these problems, it is usually up to the application's user to discover the cause of the connectivity problems and ?nd a solution. In this paper we describe SmartSockets1 a communication library that lifts this burden by automatically discovering the connectivity problems and solving them with as little support from the user as possible. Copyright 2007 ACM.
this paper presents implementation of a very fast parallel complex FFT on M2, the second generation of MorphoSys Reconfigurable computation platform, which is targeting on streamed applications such as multimedia and ...
详细信息
ISBN:
(纸本)0769520464
this paper presents implementation of a very fast parallel complex FFT on M2, the second generation of MorphoSys Reconfigurable computation platform, which is targeting on streamed applications such as multimedia and DSP. the proposed mapping comprises fast presorting, cascaded radix-2 stages, and post-reordering. Data and twiddle factors are 16-bit real and 16-bit imaginary in 2's complement format and scaling is performed to avoid overflow. the mapping is tested on our cycle-accurate simulator, "Mulate", and the performance is encouragingly better than other architectures such as Imagine and VIRAM. Moreover, the performance is scalable according to FFT sizes. Since there is no functionality specifically tailored to FFT, the results demonstrate the capability of MorphoSys architecture to extract parallelism from streamed applications. Further rationales are given based on the concepts of scalar operand networks and memory hierarchy.
Welcome to the 16th Workshop on Interaction between Compilers and computerarchitectures (INTERACT-16), held on February 25, 2012, in New Orleans, Louisiana. this year, the workshop was held in conjunction withthe 18...
详细信息
A bipartite graph G = (V, W, E) is convex if there exists an ordering of the vertices of W such that, for each v is an element of V, the neighbors of v are consecutive in W. In this work we describe a BSP/CGM algorith...
详细信息
ISBN:
(纸本)0769520464
A bipartite graph G = (V, W, E) is convex if there exists an ordering of the vertices of W such that, for each v is an element of V, the neighbors of v are consecutive in W. In this work we describe a BSP/CGM algorithm for finding a maximum matching in a convex bipartite graph. For p processors, the algorithm runs in time O((\V\/p) lg(\V\/p) lgp) and it uses O(lgp) communication rounds.
暂无评论