Scalability Up to P N1/5 Nodes for the 2D Helmholtz Equation

来源 :第八届工业与应用数学国际大会 | 被引量 : 0次 | 上传用户：lpt207

【摘要】

：

　　We present a solver for the 2D high-frequency Helmholtz equation in heterogeneous acoustic media,with online parallel complexity that scales optimally as O(

【作者】

：

DemanetLaurent;

【机构】

：

MIT

【出处】

：

第八届工业与应用数学国际大会

【发表日期】

：

2015年期

下载到本地 , 更方便阅读

下载此文赞助VIP

声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架

论文部分内容阅读

　　We present a solver for the 2D high-frequency Helmholtz equation in heterogeneous acoustic media,with online parallel complexity that scales optimally as O(N/P),where N is the number of volume unknowns,and P is the number of processors,as long as P=O(N1/5).

其他文献

Control and Observation for Stochastic Hyperbolic Equations

　　In this talk,I will explain the main difficulty for the controllability of stochastic hyperbolic equations.

会议

An Algorithm for Parametric Integer Programming

　　Parametric integer programming(PIP)is an integer programming such that the right-hand-side vector contains parameters.

会议

Developments in Computer Algebra Research and Collaboration with Industry

　　Recently,development in computer algebra research becomes important and several applications of number theory in cryptography are found.

会议

Mathematics for Industry 2:Methods from number theory,geometry,computer algebra,and integer programm

　　Rapid development of information technologies poses various challenges in the areas of information security and performance optimization.

会议

Secret Sharing Secure Against Active Adversaries with Applications to Cloud Security and Long-term S

　　Secret sharing realizes splitting of the data into pieces(called "shares")among a set of parties such that only designated subsets of them can reconstruct i

会议

Sampling High-dimensional Distributions Using Linear Iterations

　　Sampling from a Gaussian distribution is essentially the same task as solving a linear system in the precision matrix,plus drawing normal random variables.

会议

A Randomized Likelihood Method for Data Reduction in Large-scale Inverse Problems

　　We develop innovative approach to address the big data challenge in large-scale inverse problems and UQ governed by expensive PDEs.

会议

Affine Invariante MCMC's,the Way Forward or the Least We Should Ask For?

　　Affine invariant MCMC methods(eg.the 't-walk' or the 'emcee hammer')are regularly used in some areas of Bayesian inference,although still not used extensive

会议

Ensemble-based MCMC Methods for Exploring Large-scale High Dimensional Bayesian Inverse Problems

　　Sampling techniques are important for large-scale high dimensional Bayesian inferences.

会议

Advances in MCMC and related sampling methods for large-scale inverse problems

　　Inverse problems convert indirect measurements into useful characterizations of the parameters of a physical system.

会议

Scalability Up to P N1/5 Nodes for the 2D Helmholtz Equation

与本文相关的学术论文