A discussion on calculating eigenvalues of real symmetric tridiagonal matrices on a GPU
A discussion on calculating eigenvalues of real symmetric tridiagonal matrices on a GPU
复制标题
GPU上计算实对称三对角矩阵特征值的讨论
DOI：
10.1145/2082156.2082184
复制
发表时间：
2011
期刊：
影响因子：
--
通讯作者：
K.Oguri
中科院分区：
文献类型：
--
作者：
K.Matsunobu;K.Dohi;Y.Shibata;K.Oguri
虽然GPU在广泛的应用领域中作为加速器引起了人们的关注，但架构与所选算法之间的兼容性对于有效发挥其潜在性能至关重要。本文重点研究了给定真实的对称三对角矩阵的特征值计算，并比较了QR方法和二分法的GPU实现。共四个不同的GPU架构的实现，并显示和比较，以揭示算法和架构之间的亲和力。
While GPUs are attracting attention as an accelerator in wide-ranged application areas, compatibility between the architecture and selected algorithm is important to effectively bring out their potential performance. This paper focuses on eigenvalue calculation from a given real symmetric tridiagonal matrix and compares GPU implementations for the QR method and the bisection method. Implementation for a total of four different GPU architectures are shown and compared to reveal the affinity between algorithms and architectures.