Program linkage method for using dummy arguments to link program having
hierarchical structure
    2.
    发明授权
    Program linkage method for using dummy arguments to link program having hierarchical structure 失效
    使用虚拟参数来链接具有层次结构的程序的程序链接方法

    公开(公告)号:US5675806A

    公开(公告)日:1997-10-07

    申请号:US557555

    申请日:1995-11-14

    IPC分类号: G06F9/40 G06F9/445 G06F9/44

    CPC分类号: G06F9/44521

    摘要: A program linkage method comprising a step for allocating an array work area used by a low-order subprogram and an area for a table used to manage said array work area in a one-dimensional array developed for the main program of a program having hierarchical structure, a step for cataloging information used to identify said array work area to said table, and a step for retrieving information cataloged in said table for identifying said array work area used by said low-order subprogram and for passing said information as a dummy argument to said low-order subprogram.

    摘要翻译: 一种程序联动方法,包括用于分配由低阶子程序使用的阵列工作区域和用于用于管理所述阵列工作区域的区域的步骤,所述表格用于为具有分层结构的程序的主程序开发的一维阵列 用于将用于识别所述阵列工作区的信息编目到所述表的步骤,以及用于检索在所述表中编目的用于识别由所述低阶子程序使用的所述阵列工作区并将所述信息作为虚拟参数传递的信息的步骤 说低阶子程序。

    Epitaxial growth method of semiconductor crystal and molecular beam
epitaxy apparatus for the same
    3.
    发明授权
    Epitaxial growth method of semiconductor crystal and molecular beam epitaxy apparatus for the same 失效
    半导体晶体和分子束外延装置的外延生长方法相同

    公开(公告)号:US5505159A

    公开(公告)日:1996-04-09

    申请号:US341530

    申请日:1994-11-17

    CPC分类号: C30B23/02 C30B29/40

    摘要: In an epitaxial growth of a group III-V compound semiconductor crystal, there is provided a substrate on which group III element halide molecules are adsorbed. A beam of group V element hydride molecules is supplied toward the substrate for reaction of the group V element hydride and the group III element halide. The vibration energy of each of group V element hydride molecules is excited in the beam and the orientation of the group V element hydride molecules is aligned. As a result, the supplied group V atom directly combines with the group III atom.

    摘要翻译: 在III-V族化合物半导体晶体的外延生长中,提供吸附有III族元素卤化物分子的基板。 向基板供给V族元素氢化物分子束,用于V族元素氢化物和III族元素卤化物的反应。 V族元素氢化物分子中的每一个的振动能量在束中被激发,并且V族元素氢化物分子的取向被对准。 结果,所提供的V族原子与III族原子直接结合。

    Parallel calculation method and device
    4.
    发明授权
    Parallel calculation method and device 失效
    并行计算方法和装置

    公开(公告)号:US07885796B2

    公开(公告)日:2011-02-08

    申请号:US10572745

    申请日:2004-09-21

    CPC分类号: G06F17/13

    摘要: A method for executing calculation of the Hartree-Fock method in a molecular orbital method by a distributed memory parallel computing system includes the steps of: using a computer cluster made up of a plurality of computers; dividing a density matrix into multiple density submatrixes and distributing them to the individual computers and storing therein; and executing calculation processes such as two-electron integration or the like relating to density submatrixes in each computer while sequentially transferring the multiple density submatrixes between the multiple computers.

    摘要翻译: 一种通过分布式存储器并行计算系统以分子轨道方式执行Hartree-Fock方法的计算的方法包括以下步骤:使用由多台计算机构成的计算机群; 将密度矩阵划分为多密度子矩阵并将其分配到各个计算机并存储在其中; 并且在多个计算机之间顺序传送多个密度子矩阵的同时,在每台计算机中执行与密度子矩阵相关的两电子积分等计算处理。

    Molecular simulation method and device
    6.
    发明申请
    Molecular simulation method and device 审中-公开
    分子模拟方法及装置

    公开(公告)号:US20070043545A1

    公开(公告)日:2007-02-22

    申请号:US10573023

    申请日:2004-09-22

    IPC分类号: G06G7/48 G06G7/58

    CPC分类号: G16C10/00 G16C20/50

    摘要: A molecular simulation method for dividing a molecule or a part of molecule to be simulated into a QM space and an MM apace and applying an ab initio molecular orbital method to the QM space and a method based on an empirical potential to the MM space to perform molecular simulation includes the steps of: retrieving structure data on the molecule or part of molecule to be simulated from a storage unit, and dividing the structure data into the QM space and the MM space; and replacing a part of a total energy expression in the ab initio molecular orbital method concerning the QM space with an empirical potential.

    摘要翻译: 一种分子模拟方法,用于将待模拟的分子或部分分子划分为QM空间和MM格子,并将起始分子轨道方法应用于QM空间,以及基于MM空间的经验潜力来执行的方法 分子模拟包括以下步骤:从存储单元检索要模拟的分子或部分分子上的结构数据,并将结构数据划分为QM空间和MM空间; 并以具有经验潜力的关于QM空间的从头分子轨道方法替代总能量表达的一部分。

    Parallel calculation method and device
    7.
    发明申请
    Parallel calculation method and device 失效
    并行计算方法和装置

    公开(公告)号:US20060271301A1

    公开(公告)日:2006-11-30

    申请号:US10572745

    申请日:2004-09-21

    IPC分类号: G06G7/58 G06F19/00

    CPC分类号: G06F17/13

    摘要: A method for executing calculation of the Hartree-Fock method in a molecular orbital method by a distributed memory parallel computing system includes the steps of: using a computer cluster made up of a plurality of computers; dividing a density matrix into multiple density submatrixes and distributing them to the individual computers and storing therein; and executing calculation processes such as twoelectron integration or the like relating to density submatrixes in each computer while sequentially transferring the multiple density submatrixes between the multiple computers.

    摘要翻译: 一种通过分布式存储器并行计算系统以分子轨道方式执行Hartree-Fock方法的计算的方法包括以下步骤:使用由多台计算机构成的计算机群; 将密度矩阵划分为多密度子矩阵并将其分配到各个计算机并存储在其中; 并且在多个计算机之间顺序地传送多个密度子矩阵的同时,在各计算机中执行与密度子矩阵相关的两电子积分等计算处理。

    Parallel computing method for total energy and energy gradient of non experience molecular-orbital method
    8.
    发明授权
    Parallel computing method for total energy and energy gradient of non experience molecular-orbital method 失效
    非经验分子轨道法的总能量和能量梯度的并行计算方法

    公开(公告)号:US07343277B2

    公开(公告)日:2008-03-11

    申请号:US10490703

    申请日:2002-09-25

    IPC分类号: G06G9/44

    摘要: A parallel computing method by using a parallel computer having a plurality of processors is provided, wherein when a 2-electron integration is transformed from an atomic-orbital base (rs|tu) to a molecular-orbital base (ab|cd), indexes r and s of an atomic orbital to be calculated are designated for the plurality of processors and each of the plurality of processors performs processing for all combinations of designated indexes R and S, and indexes t and u of the atomic orbital to be calculated. Subsequently, the grain sizes are equalized, many commodity processors can be connected with high performance, and the cost of a high-speed operation computer reduces. Further, the computer can obtain a large main-memory area, as a whole, through the use of inexpensive local memories.

    摘要翻译: 提供了一种通过使用具有多个处理器的并行计算机的并行计算方法,其中当从原子轨道基(rs | tu)到分子轨道基(ab | cd)转换2电子积分时, 对于多个处理器指定要计算的原子轨道的r和s,并且多个处理器中的每个处理器对指定索引R和S的所有组合以及要计算的原子轨道的索引t和u执行处理。 随后,晶粒尺寸相等,许多商品处理器可以高性能地连接,并且高速运行计算机的成本降低。 此外,计算机可以通过使用便宜的本地存储器整体地获得大的主存储区域。