A wavelet-based parallel implementation is presented for image encoding on a multi-DSP system. the implementation is utilizing the discrete wavelet transform (DWT) and is realized in parallel processor architecture. T...
详细信息
ISBN:
(纸本)9780780397361
A wavelet-based parallel implementation is presented for image encoding on a multi-DSP system. the implementation is utilizing the discrete wavelet transform (DWT) and is realized in parallel processor architecture. the implementation has a very flexible architecture, which allows addition of extra slave processors (SPs) to the system whenever more computational power is needed. Performance of the implementation is measured and compared to a sequential reference implementation. Experimental results show that the parallel implementation is very efficient and overpowers the sequential counterpart considerably.
In tins impel we discuss the implementation of a simple pedestrian simulation that, uses a multi agent based design pattern developed by the CoSMoS research group Given the nature of Multi Agent Systems (MAS), paralle...
详细信息
ISBN:
(纸本)9783642144028
In tins impel we discuss the implementation of a simple pedestrian simulation that, uses a multi agent based design pattern developed by the CoSMoS research group Given the nature of Multi Agent Systems (MAS), parallelprocessing techniques are inevitably used in their implementation Most, of these approaches rely on conventional parallel programming techniques;such as threads, Message Passing Interface (MPI) and Remote Method Invocation (RMI) the CoSMoS design pattern us are founded on the use of Communicating Sequential Processes (CSID) a pal and computing paradigm that emphasises a process or rather than object oriented programming perspective
In this work we present the implementation of an application to simulate the evolution of pressure and temperature inside a cavity when acoustic energy is injected, a physical system currently under intensive research...
详细信息
ISBN:
(纸本)9783642143892
In this work we present the implementation of an application to simulate the evolution of pressure and temperature inside a cavity when acoustic energy is injected, a physical system currently under intensive research. the particular features of the equations of the model makes the simulation problem very stiff and time consuming. However, intrinsic parallelism makes the application suitable for implementation in GPUs providing the researchers with a very useful tool to study the problem at a very reasonable price. In our experiments the problem was solved in less than half the time required by CPUs.
In this paper we deal withthe problem of finding an optimal query execution plan in database systems. We improve the analysis of a polynomial-time approximation algorithm due to Makino et al. for designing query exec...
详细信息
In this paper we deal withthe problem of finding an optimal query execution plan in database systems. We improve the analysis of a polynomial-time approximation algorithm due to Makino et al. for designing query execution plans with almost optimal number of parallel steps. this algorithm is based on the concept of edge ranking of graphs. We use a new upper bound for the edge ranking number of a tree to derive a better worst-case performance guarantee for this algorithm. We also present some experimental results obtained during the tests of the algorithm on random graphs in order to compare the quality of both approximation ratios on average. Boththeoretical analysis and experimental results indicate the superiority of our approach.
Nowadays GPUs become extremely promising multi/many-core architectures for a wide range of demanding applications. Basic features of these architectures include utilization of a large number of relatively simple proce...
详细信息
ISBN:
(纸本)9783642143892
Nowadays GPUs become extremely promising multi/many-core architectures for a wide range of demanding applications. Basic features of these architectures include utilization of a large number of relatively simple processing units which operate in the SIMD fashion, as well as hardware supported, advanced multithreading. However, the utilization of GPUs in an every-day practice is still limited, mainly because of necessity of deep adaptation of implemented algorithms to a target architecture. hi this work, we propose how to perform such an adaptation to achieve an efficient parallel implementation of the conjugate gradient (CG) algorithm, which is widely used for solving large sparse linear systems of equations, arising e.g. in FEM problems. Aiming at efficient;implementation of the main operation of the CG algorithm, which is sparse matrix-vector multiplication (SpMV), different techniques of optimizing access to the hierarchical memory of GPUs are proposed and studied. the experimental investigation of a proposed CUDA-based implementation of the CG algorithm is carried out On two CPU architectures: GeForce 8800 and Tesla C1060. It has been shown that optimization of access to CPU memory allows us to reduce considerably the execution time of the SpMV operation, and consequently to achieve a significant speedup over CPUs when implementing the whole CC algorithm.
Application of FPGA (Field Programmable Gate Array) to a numerical simulator for the purpose of achieving fast computation of MDOF structural model in the real-time experimental hybrid simulation system is investigate...
详细信息
ISBN:
(纸本)9789076019314
Application of FPGA (Field Programmable Gate Array) to a numerical simulator for the purpose of achieving fast computation of MDOF structural model in the real-time experimental hybrid simulation system is investigated. Design and performance analysis of logic circuits required for high-speed dynamic simulation of MDOF structural system model are performed, and verification of the implementation of the designed logic circuit to actual FPGA device is verified. Furthermore, limitation of computational performance of a single FPGA unit in terms of the speed of the processing and the number of DOF of the structural model applied is investigated.
this paper presents a fine-grained parallelprocessing technique applied to the transient and steady state analysis of electric power systems containing linear, nonlinear and time varying elements. this approach is ba...
详细信息
Methods to accurately measure Phase-locked loop lock time in multisite production environment has been presented and explained. the methods are applicable for testing transceiver frequency settling times, and frequenc...
详细信息
ISBN:
(纸本)9780780397361
Methods to accurately measure Phase-locked loop lock time in multisite production environment has been presented and explained. the methods are applicable for testing transceiver frequency settling times, and frequency and phase errors after settling for multiple devices under test in parallel using on board frequency mixers and RF signal generators or using RF receivers of automated testers. Inverse FFT was used to measure the PLL lock time in a case when PLL frequency error exists.
the considerable computation time of a practical application of sequential algorithms for simulating thermal and flow distribution in pipe networks is the motivating factor to study their parallel implementation the m...
详细信息
ISBN:
(纸本)9783642144028
the considerable computation time of a practical application of sequential algorithms for simulating thermal and flow distribution in pipe networks is the motivating factor to study their parallel implementation the mathematical model formulated and studied in the paper requires the solution of a set of nonlinear equations, which are solved by the Newton-Raphson method An object-oriented solver automatically formulates the equations for networks of an arbitrary topology the hydraulic model that is chosen as a benchmark consists of nodal flows and loop equations A general decomposition algorithm for analysis of flow and temperature distribution in a pipe network is presented, and results of speedup of its parallel implementation are demonstrated
Stochastic simulation of biological systems becomes widely used, since it can intuitively account for the fluctuation of species with a few number of molecules. However, for bigger models and/or models with mixed abun...
详细信息
ISBN:
(纸本)9781509029877
Stochastic simulation of biological systems becomes widely used, since it can intuitively account for the fluctuation of species with a few number of molecules. However, for bigger models and/or models with mixed abundance of molecules, stochastic simulation fails to produce the required results in reasonable time. parallel simulation can offer a solution for this challenge. Nevertheless, currently available parallel software tools either provide a coarse-grained parallelization or a general-purpose fine-grained parallel simulation of the wellknown stochastic simulation algorithm (SSA). the former can only take advantage of parallelprocessing if multiple runs have to be performed, while the latter requires extensive synchronization and communication between the different processing nodes each time a reaction is to fire. In this paper, a fine-grained parallelization approach is presented that takes advantage of the underlying model semantics to improve the simulator performance. the proposed method is applied to the yeast cell cycle regulation, which is an example of biological models that requires extensive investigation.
暂无评论