The proceedings contain 52 papers. The special focus in this conference is on Robust Preconditioning algorithms, Monte Carlo Methods, Advanced programming Environments and Large-Scale Computations in Air Pollution Mod...
ISBN:
(纸本)3540430431
The proceedings contain 52 papers. The special focus in this conference is on Robust Preconditioning algorithms, Monte Carlo Methods, Advanced programming Environments and Large-Scale Computations in Air Pollution Modelling. The topics include: Optimizing two-level preconditionings for the conjugate gradient method;on the parallelization of the sparse grid approach for data mining;java communications for large-scale parallel computing;iterative aggregation/disaggregation methods for computing some characteristics of Markov chains;time-integration algorithms for the computer treatment of the horizontal advection in air pollution models;sobolev space preconditioning for mixed nonlinear elliptic boundary value problems;on a schur complement approach for solving two-level finite element systems;on-line state estimation of maneuvering objects by sequential Monte Carlo algorithm;an improved Monte Carlo algorithm for elastic electron backscattering from surfaces;a quasi-Monte Carlo method for integration with improved convergence;solving systems of linear algebraic equations using quasirandom numbers;Monte Carlo analysis of the small-signal response of charge carriers;interactive control and debugging of distribution;program development environment for openMP programs on ccNUMA architectures;object-oriented framework for large scale air pollution modeling;evaluation and reliability of meso-scale air pollution simulations;fine-grid resolution in Danish eulerian model and an implementation on SGI origin 2000 computer;modelling framework for atmospheric mercury over the Mediterranean region;iterative load balancing schemes for air pollution models;long-term estimates of sulfur deposition in the region of southeastern Europe and handling systems from non-linear theory of elasticity.
Two classic problems that are computationally intensive and show good speedup and scalability when solved in a parallelprogramming environment are used to test the different resource allocation and management algorit...
详细信息
ISBN:
(纸本)0769514324
Two classic problems that are computationally intensive and show good speedup and scalability when solved in a parallelprogramming environment are used to test the different resource allocation and management algorithms used with the node intrusion and failure experiment. We divide the adaptive resource allocation experiments into two groups: (i) automatic survivability and scalability (ii) assessment of real-time quality of service (QoS). In the former, we use different algorithms to detect failed programs, host and network resources and idle times, computing an allocation, enactment of an allocation, and restart notification. We also use different techniques to detect dynamic paths that are receiving poor QoS possibly due to overload and to "scale up" such paths via reallocation. In the latter case, we use different fitness functions to classify the connections and the resources available on the nodes and study the effects of these on the overall resource allocation and the eventual speedup.
Reductions are important and time-consuming operations in many scientific codes. Effective parallelization of reductions is a critical transformation for loop parallelization, especially for sparse, dynamic applicatio...
详细信息
Reductions are important and time-consuming operations in many scientific codes. Effective parallelization of reductions is a critical transformation for loop parallelization, especially for sparse, dynamic applications. Unfortunately, conventional reduction parallelization algorithms are not scalable. In this paper, we present new architectural support that significantly speeds up parallel reduction and makes it scalable in shared-memory multiprocessors. The required architectural changes are mostly confined to the directory controllers. Experimental results based on simulations show that the proposed support is very effective. While conventional software-only reduction parallelization delivers average speedups of only 2.7 for 16 processors, our scheme delivers average speedups of 7.6.
The proceedings contain 35 papers. The special focus in this conference is on Formal Methods Europe. The topics include: Lightweight formal methods;a way to combine dynamic properties and b refinement;mechanized analy...
ISBN:
(纸本)3540417915
The proceedings contain 35 papers. The special focus in this conference is on Formal Methods Europe. The topics include: Lightweight formal methods;a way to combine dynamic properties and b refinement;mechanized analysis of behavioral conformance in the eiffel base libraries;proofs of correctness of cache-coherence protocols;model-checking over multi-valued logics;LTL model checking of CSP by refinement;avoiding state explosion for distributed systems with timestamps;secrecy-preserving refinement;a rigorous approach to modeling and analyzing e-commerce architectures;a programming model for wide-area computing;validation of UML models thanks to Z and lustre;components, contracts, and connectors for the unified modelling language UML;an integrated approach to specification and validation of real-time systems;real-time logic revisited;improvements in BDD-based reachability analysis of timed automata;serialising parallel processes in a hardware/software partitioning context;verifying implementation relations;faithful translations among models and specifications;composing contracts: an adventure in financial engineering;from complex specifications to a working prototype;coverage directed generation of system-level test cases for the validation of a DSP system;using formal verification techniques to reduce simulation and test effort;transacted memory for smart cards;a heuristic for symmetry reductions with scalarsets;view updatability based on the models of a formal specification;grammar adaptation;test-case calculation through abstraction;a modular approach to the specification and validation of an electrical flight control system and a combined testing and verification approach for software reliability.
A novel parallel architecture for estimating computationally intensive 4th-order cumulants is presented. Different from most systolic array implementations, a MIMD array processor is used to efficiently compute the cu...
详细信息
ISBN:
(纸本)0780366859
A novel parallel architecture for estimating computationally intensive 4th-order cumulants is presented. Different from most systolic array implementations, a MIMD array processor is used to efficiently compute the cumulants by exploiting the algorithmic parallelism, reducing operand-fetching operations and by optimising the processing elements' architectural design. It is shown that by breaking the algorithm into a number of separate stages and reorganising all computations in matrix block form, a significant computational speed-up can be obtained, which increases the applicability of cumulant-based algorithms in a real-time system.
The emergence of scalable computer architectures using clusters of PCs (or PC-SMPs) with commodity networking has made them attractive platforms for high-end scientific computing. Currently, message passing (MP) and s...
详细信息
ISBN:
(纸本)0769509908
The emergence of scalable computer architectures using clusters of PCs (or PC-SMPs) with commodity networking has made them attractive platforms for high-end scientific computing. Currently, message passing (MP) and shared address space (SAS) are the two leading programming paradigms for these systems. MP has been standardized with MPI, and is the most common and mature parallelprogramming approach. However, MP code development can be extremely difficult, especially for irregularly structured computations. SAS offers substantial ease of programming, but may suffer from performance limitations due to poor spatial locality and high protocol overhead. In this paper, we compare the performance of and programming effort required for six applications under both programming models on a 32-CPU PC-SMP cluster. Our application suite consists of codes that typically do not exhibit scalable performance under shared-memory programming due to their high communication-to-computation ratios and complex communication patterns. Results indicate that SAS can achieve about half the parallel efficiency of MPI for most of our applications; however on certain classes of problems, SAS performance is competitive with MPI.
At present,the scale of geotechnical projects becomes larger and *** structures and geological environments met in the projects are more and more complicated and the requirements for computing precision and time becom...
详细信息
At present,the scale of geotechnical projects becomes larger and *** structures and geological environments met in the projects are more and more complicated and the requirements for computing precision and time become higher and *** satisfy these increasing demands,we propose parallel computing based on PC on Windows NT for geotechnical engineering problems. Some algorithms on parallel finite element method,parallel boundary element method,parallel finite difference method and parallel genetic programming are *** results of slope stability evaluation by parallel genetic programming show that the parallel system on Windows NT can overcome the difficulties of large-scale geotechnical simulations.
暂无评论