专利快速检索-快速检索全球专利，免费商用专利数据库-IPRDB

1. 发明申请

US20150134714A1 SHARED MEMORY EIGENSOLVER 有权
标题翻译：共享内存EIGENSOLVER
公开(公告)号：US20150134714A1
公开(公告)日：2015-05-14
申请号：US14537839
申请日：2014-11-10
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F17/16
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： Disclosed herein is a shared memory systems that use a combination of SBR and MRRR techniques to calculate eigenpairs for dense matrices having very large numbers of rows and columns. The disclosed system allows for the use of a highly scalable tridiagonal eigensolver. The disclosed system likewise allows for allocating a different number of t00hreads to each of the different computational stages of the eigensolver.
摘要翻译：这里公开了一种共享存储器系统，其使用SBR和MRRR技术的组合来计算具有非常大数量的行和列的密集矩阵的特征对。所公开的系统允许使用高度可缩放的三角形固定器。所公开的系统同样允许将不同数量的t00hreads分配给固定器的不同计算阶段。

2. 发明申请

US20160299874A1 SHARED MEMORY EIGENSOLVER 审中-公开
标题翻译：共享内存EIGENSOLVER
公开(公告)号：US20160299874A1
公开(公告)日：2016-10-13
申请号：US15132085
申请日：2016-04-18
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F17/16
CPC分类号： G06F17/16 , G06F12/0207
摘要： Disclosed herein is a shared memory system that use a combination of SBR and MRRR techniques to calculate eigenpairs for dense matrices having very large numbers of rows and columns. The disclosed system allows for the use of a highly scalable tridiagonal eigensolver. The disclosed system likewise allows for allocating a different number of threads to each of the different computational stages of the eigensolver.
摘要翻译：这里公开了一种共享存储器系统，其使用SBR和MRRR技术的组合来计算具有非常大数量的行和列的密集矩阵的特征对。所公开的系统允许使用高度可缩放的三角形固定器。所公开的系统同样允许将不同数量的线程分配给固定器的不同计算阶段。

3. 发明申请

US20140331014A1 Scalable Matrix Multiplication in a Shared Memory System 审中-公开
标题翻译：在共享内存系统中的可扩展矩阵乘法
公开(公告)号：US20140331014A1
公开(公告)日：2014-11-06
申请号：US14041974
申请日：2013-09-30
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F12/00
CPC分类号： G06F15/17306 , G06F17/16
摘要： High performance computing systems perform complex or data-intensive calculations using a large number of computing nodes and a shared memory. Disclosed methods and systems provide nodes having a special-purpose coprocessor to perform these calculations, along with a general-purpose processor to direct the calculations. Computational data transfer from the shared memory to the coprocessor incurs a data copying latency. To reduce this latency as experienced by the coprocessor, a complex computation is divided into work units, and one or more threads executing on the processor copy the work units from the shared memory to a local buffer memory of a computing node. By buffering these data for transfer from the local memory to coprocessor memory, and by ensuring that new data are copied while the coprocessor operates on older data, data copying latency is hidden from the coprocessor.
摘要翻译：高性能计算系统使用大量计算节点和共享内存执行复杂或数据密集型计算。公开的方法和系统提供具有专用协处理器的节点来执行这些计算，以及用于指导计算的通用处理器。从共享存储器到协处理器的计算数据传输引起数据复制延迟。为了减少由协处理器所经历的延迟，复杂的计算被划分为工作单元，并且在处理器上执行的一个或多个线程将工作单元从共享存储器复制到计算节点的本地缓冲存储器。通过缓冲这些数据从本地存储器转移到协处理器存储器，并且通过确保在协处理器对旧数据进行操作时复制新数据，数据复制等待时间将从协处理器中隐藏起来。

4. 发明授权

US09798594B2 Shared memory eigensolver 有权
公开(公告)号：US09798594B2
公开(公告)日：2017-10-24
申请号：US15408092
申请日：2017-01-17
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F17/16 , G06F9/54 , G06F9/50 , G06F9/48
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： Disclosed herein is a shared memory systems that use a combination of SBR and MRRR techniques to calculate eigenpairs for dense matrices having very large numbers of rows and columns. The disclosed system allows for the use of a highly scalable tridiagonal eigensolver. The disclosed system likewise allows for allocating a different number of threads to each of the different computational stages of the eigensolver.

5. 发明申请

US20170228267A1 SHARED MEMORY EIGENSOLVER 有权
公开(公告)号：US20170228267A1
公开(公告)日：2017-08-10
申请号：US15408092
申请日：2017-01-17
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F9/54 , G06F9/48 , G06F9/50 , G06F17/16
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： Disclosed herein is a shared memory systems that use a combination of SBR and MRRR techniques to calculate eigenpairs for dense matrices having very large numbers of rows and columns. The disclosed system allows for the use of a highly scalable tridiagonal eigensolver. The disclosed system likewise allows for allocating a different number of threads to each of the different computational stages of the eigensolver.

6. 发明授权

US09547882B2 Shared memory eigensolver 有权
标题翻译：共享记忆体分子
公开(公告)号：US09547882B2
公开(公告)日：2017-01-17
申请号：US14537839
申请日：2014-11-10
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06T1/60 , G06F17/16 , G06F12/08 , G06F9/54 , G06F9/50
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： Disclosed herein is a shared memory systems that use a combination of SBR and MRRR techniques to calculate eigenpairs for dense matrices having very large numbers of rows and columns. The disclosed system allows for the use of a highly scalable tridiagonal eigensolver. The disclosed system likewise allows for allocating a different number of threads to each of the different computational stages of the eigensolver.
摘要翻译：这里公开了一种共享存储器系统，其使用SBR和MRRR技术的组合来计算具有非常大数量的行和列的密集矩阵的特征对。所公开的系统允许使用高度可缩放的三角形固定器。所公开的系统同样允许将不同数量的线程分配给固定器的不同计算阶段。

7. 发明授权

US09262799B2 Shared memory eigensolver 有权
标题翻译：共享记忆体分子
公开(公告)号：US09262799B2
公开(公告)日：2016-02-16
申请号：US14536477
申请日：2014-11-07
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06F15/80 , G06F15/16 , G06T1/60 , G06F17/16 , G06F12/08 , G06F9/54
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： A method for computing eigenvectors and eigenvalues of a square matrix in a high performance computer involves dynamically reallocating the computer's computing cores for various phases of the computation process.
摘要翻译：用于计算高性能计算机中的方阵的特征向量和特征值的方法涉及对计算过程的各个阶段动态地重新分配计算机的计算核心。

8. 发明申请

US20150130825A1 Shared Memory Eigensolver 有权
标题翻译：共享内存Eigensolver
公开(公告)号：US20150130825A1
公开(公告)日：2015-05-14
申请号：US14536477
申请日：2014-11-07
申请人： Silicon Graphics International Corp.
发明人： Cheng Liao
IPC分类号： G06T1/60
CPC分类号： G06F9/544 , G06F9/4887 , G06F9/5016 , G06F9/5066 , G06F12/08 , G06F17/16 , G06F2212/2542 , G06T1/60
摘要： A method for computing eigenvectors and eigenvalues of a square matrix in a high performance computer involves dynamically reallocating the computer's computing cores for various phases of the computation process.
摘要翻译：用于计算高性能计算机中的方阵的特征向量和特征值的方法涉及对计算过程的各个阶段动态地重新分配计算机的计算核心。

你已经成功收藏专利！

检索式保存成功!

IPRDB

热门服务

关于我们

友情链接

联系方式