搜索资源列表
mpipar.rar
- 在MPI上实现的矩阵相乘并行计算的源程序。
matmult
- 并行矩阵乘法。用于mpi和openMP混合并行求解矩阵乘法问题。适用于分布共享存储cluster环境。无解压密码。-parallel matrix multiplication. OpenMP for mixed MPI and parallel matrix multiplication to solve the problem. Distribution applicable to the shared storage cluster environment. Without extract
invert
- 矩阵求逆的并行算法,使用高斯约当消去法,是从一参考书中copy过来的,谢谢原作者。-matrix inversion of the parallel algorithm, about when the use of Gaussian Elimination Act, a reference book from which the copy over, Thank you, the original author.
superlu_mt_1.0.tar
- 可以用在并行计算中的代码,支持稀疏矩阵的LU分解,用来求方程组的解-can be used in parallel computing code, sparse matrix support the LU decomposition, used for the solution of the equations
128bitminus
- 128乘法模拟器 c M位乘N位不带符号整数的阵列乘法中加法---移位操作的被加数矩阵.每一部分乘积项ab叫做一个被加数.m*n个被加数可以用m*n个”与门”并行的产生. 以5位乘5位不带符号的阵列乘法器(m=n=5)为例(如下图): FA为一位全加器,FA的斜线方向为进位输出,竖线方向为和输出,而所有被加数项的排列和正常的A*B=P乘法过程中的被加数矩阵相同.图中用矩形围成的阵列中最后一行构成一个行波进位加法器,其时间延迟为(n-1)2T.当然,为了缩短加法时间,最后一行的行
18-4
- cannon矩阵并行算法,mpi编写,计算过程中要通过消息传递完成不同节点间的通信-cannon matrix parallel algorithms, mpi writing, the calculation process should be completed by the message-passing communication between different nodes
MultiThreadMutiple
- 矩阵并行乘法 java多线程实现-matrix inversion algorithm
mpiPopenmp
- 采用MPI+OpenMP混合模型设计并实现光化学反应双层并行模拟算法,上层基于MPI实现节点间的原子分解并行,下层基于OpenMP实现节点内的多线程矩阵并行乘法。-With MPI+OpenMP mixed model was designed and implemented The photochemical reaction bunk parallel simulation algorithms, the upper atomic decomposition based on inter-n
cuda2
- 最简单的矩阵并行计算方法,很简单 调用CUDA计算得到的 非常简单-The easiest parallel matrix calculation method is very simple call to get a very simple calculation CUDA
mm
- 多线程实现矩阵并行相乘,在实现方式上采用多线程实现相乘-Implementation of parallel matrix multiplication of multi thread
cuda+mpi
- 并行计算中矩阵相乘的mpi和cuda两种方法,已完成检验,有结果输出(In parallel computing, two methods of matrix multiplication, MPI and CUDA, have been tested and output with results)
mpi_mat_vect_time
- 使用mpi编写的进行矩阵乘法运算的并行程序。(An parallel program using MPI to realize the multiplication of matrices.)
colt-1.2.0
- Colt是一个高性能的数学库,由以下几个子库构成: Colt库:基本的动态数组、稀疏矩阵、线性代数。 Jet库:数理统计、直方图。 CoreJava库:类printf的打印函数,并行计算。(Colt is a high performance math library composed of the following sub libraries: Colt libraries: basic dynamic arrays, sparse matrices, linear algebra.
Complex_matrix_row_col
- 通过并行实现,提高复矩阵与复矩阵相乘运算速度(Increasing the speed of multiplication of complex matrices and complex matrices)
一种基于MapReduce的粗糙集并行属性约简算法
- 云计算技术是海量数据挖掘的一种高效解决方案,将MapReduce 并行计算模型与粗糙集属性约简算法相结合,提出一种基于MapReduce 的浓缩布尔矩阵并行属性约简算法。该算法提高了粗糙集属性约简算法对大数据的处理能力和效率,并能适应云计算环境。实验结果表明,所提算法具有良好的效率、加速比和可扩展性。(Cloud computing technology is a high efficient solution for massive data mining.)
115010053_Lu Danlin
- mpi non-blocking 计算矩阵(mpi non-blocking multiply matrix)
cannon
- 矩阵乘Cannon并行算法的MPI+Fortran实现(Parallel program of Cannon algorithm with MPI and Fortran)
matmul.tar
- 矩阵乘法的多线程并行实现,包括:simple,采用openmp对索引顺序ijk(一般写法)的循环实现并行;opt,采用openmp对索引顺序ikj(向量化友好)的循环实现并行;mkl,调用高性能数学库实现矩阵乘法。(Multi-thread parallelization of matrix-matrix multiplication: "simple" case, ijk loop (usual pattern); "opt" case, ikj (ve
GPU高性能编程CUDA实战—示例代码
- 结合GPU高性能编程,提供实战样例程序,包含矩阵乘法,原子操作,热传导,多GPU,多流的代码(Combined with GPU high performance programming, it provides actual sample program, including matrix multiplication, atomic operation, heat conduction, multi GPU, and multi stream code.)
code
- 平行计算课程,并行程序转置矩阵,使用3种方法:线性转置,块棋盘划分,直角划分法(Parallel computing course, parallel program transpose matrix, using three methods: linear transpose, block chessboard division, right angle division)