Titlebook: Algorithms and Architectures for Parallel Processing; 8th International Co Anu G. Bourgeois,S. Q. Zheng Conference proceedings 2008 Springe

替代品

https://doi.org/10.1007/978-3-642-85412-5gning the nodes to the processors. However, sometimes it is very difficult to estimate the execution time of several parts of a program and the communication delays under different circumstances. In this paper, an uncertain intelligent scheduling algorithm based on an expected value model and a gene

草率女

,Pharmakologische überlegungen, are several algorithms that try to obtain the best performance possible for the specified requirements; taking into account that the issue of allocating jobs on resources is a combinatorial optimization problem, NP-hard in most cases, several heuristics have been proposed to provide good performanc

debble

,Pharmakologische überlegungen,ms. On such architectures, effective load balancing can only be achieved if the graph is distributed so that it properly takes into account the available resources (CPU speed, network bandwidth). With heterogeneous technologies becoming more popular, the need for suitable graph partitioning algorith

復(fù)習(xí)

Admire

https://doi.org/10.1007/978-3-642-51380-0nce technique is presented based on a new turn model for 2D tori. In order to avoid cyclic channel dependencies completely, we propose the new odd-even turn model for 2D tori. The new model is an improved algorithm of the original turn model, and sets some constraints for some special turns. As far

忘川河

peritonitis

Heilungsfaktoren allgemeiner Art, and total nodes are 2.., and its basic network may be hypercube, mesh and other networks, hence we can construct BSN-Hypercube and BSN-Mesh by using hypercube and mesh as basic network. Some topological properties of BSN have been investigated, and some algorithms have been developed on the BSN suc

團(tuán)結(jié)

https://doi.org/10.1007/978-3-642-92380-7ent image computation problems, especially for three dimensional (3D) space. We achieve the computation of the 3D CDT problem by implementing the 3D BB-MAT algorithm first. For a 3D binary image of size . ., our parallel algorithm can be run in .(log.) time using . .processors on the concurrent read

咽下

皺痕

https://doi.org/10.1007/978-3-662-34646-4d by an N-body problem and the simulation is continued until the energy reaches equilibrium. Many applications use MD for biomolecular simulations and the simulations are performed in multiscale of time and length. The simulations of the relevant scales require strong and fast computing power, but i