会员体验
专利管家(专利管理)
工作空间(专利管理)
风险监控(情报监控)
数据分析(专利分析)
侵权分析(诉讼无效)
联系我们
交流群
官方交流:
QQ群: 891211   
微信请扫码    >>>
现在联系顾问~
热词
    • 21. 发明申请
    • Sparse vectorization without hardware gather / scatter
    • 稀疏矢量化无硬件收集/散射
    • US20080092125A1
    • 2008-04-17
    • US11549172
    • 2006-10-13
    • Roch Georges ArchambaultGeorge ChochiaPeng Zhao
    • Roch Georges ArchambaultGeorge ChochiaPeng Zhao
    • G06F9/45
    • G06F8/447
    • A target operation in a normalized target loop, susceptible of vectorization and which may, after compilation into a vectorized form, seek to operate on data in nonconsecutive physical memory, is identified in source code. Hardware instructions are inserted into executable code generated from the source code, directing a system that will run the executable code to create a representation of the data in consecutive physical memory. A vector loop containing the target operation is replaced, in the executable code, with a function call to a vector library to call a vector function that will operate on the representation to generate a result identical to output expected from executing the vector loop containing the target operation. On execution, a representation of data residing in nonconsecutive physical memory is created in consecutive physical memory, and the vectorized target operation is applied to the representation to process the data.
    • 标准化目标循环中的目标操作,易于向量化,并且可以在编译成向量化形式之后寻求对非连续物理存储器中的数据进行操作,在源代码中被识别。 硬件指令被插入到从源代码生成的可执行代码中,指示将运行可执行代码的系统在连续的物理内存中创建数据的表示。 包含目标操作的向量循环在可执行代码中被替换为对向量库的函数调用,以调用将在表示上操作的向量函数,以生成与执行包含目标的向量循环所期望的输出相同的结果 操作。 在执行时,在连续物理存储器中创建驻留在非连续物理存储器中的数据的表示,并且向量化的目标操作被应用于表示以处理数据。
    • 23. 发明申请
    • Workflow modeling using an acyclic directed graph data structure
    • 使用非循环有向图数据结构的工作流建模
    • US20050192783A1
    • 2005-09-01
    • US11120588
    • 2005-05-03
    • Garr LystadPeng Zhao
    • Garr LystadPeng Zhao
    • G06Q10/00H04J3/14
    • G06Q10/0633G06Q10/06312G06Q10/10
    • A process for modeling at least a portion of a workflow includes accessing a computer data structure to represent an acyclic directed graph (10) including multiple nodes (12) and one or more edges (14), each edge (14) linking two adjacent nodes (12). The value of a function at a selected node (12) is requested, the value of the function at the selected node (12) depending on values of the function at one or more adjacent nodes (12) lying in a first direction from the selected node (12). If a cached value of the function at the selected node (12) is not assured to be valid, then the value of the function at the selected node (12) is recomputed based on the values of the function at the one or more adjacent nodes (12) and then returned. If the cached value is assured to be valid, then the cached value is returned without recomputing the value of the function at the selected node (12).
    • 用于建模工作流的至少一部分的过程包括访问计算机数据结构以表示包括多个节点(12)和一个或多个边缘(14)的非循环有向图(10),每个边缘(14)链接两个相邻节点 (12)。 请求在所选节点(12)处的函数的值,取决于所选节点(12)处的功能的值,其取决于位于与所选择的节点(12)的第一方向上的一个或多个相邻节点(12)处的功能的值 节点(12)。 如果在所选择的节点(12)处的功能的缓存值不被确定为有效,则基于在一个或多个相邻节点处的函数的值来重新计算所选节点(12)处的功能的值 (12)然后返回。 如果缓存值被确定为有效,则返回缓存的值,而不重新计算所选节点(12)处的函数值。
    • 26. 发明授权
    • Analyze and reduce number of data reordering operations in SIMD code
    • 分析和减少SIMD代码中数据重排序的数量
    • US08954943B2
    • 2015-02-10
    • US11340452
    • 2006-01-26
    • Alexandre E. EichenbergerKai-Ting Amy WangPeng WuPeng Zhao
    • Alexandre E. EichenbergerKai-Ting Amy WangPeng WuPeng Zhao
    • G06F9/45G06F15/00G06F15/76
    • G06F8/443
    • A method for analyzing data reordering operations in Single Issue Multiple Data source code and generating executable code therefrom is provided. Input is received. One or more data reordering operations in the input are identified and each data reordering operation in the input is abstracted into a corresponding virtual shuffle operation so that each virtual shuffle operation forms part of an expression tree. One or more virtual shuffle trees are collapsed by combining virtual shuffle operations within at least one of the one or more virtual shuffle trees to form one or more combined virtual shuffle operations, wherein each virtual shuffle tree is a subtree of the expression tree that only contains virtual shuffle operations. Then code is generated for the one or more combined virtual shuffle operations.
    • 提供了一种用于分析单发多数据源代码中的数据重排序操作并从中生成可执行代码的方法。 收到输入。 识别输入中的一个或多个数据重排序操作,并将输入中的每个数据重排序操作抽象为相应的虚拟随机播放操作,使得每个虚拟随机播放操作形成表达式树的一部分。 通过将所述一个或多个虚拟随机播放树中的至少一个中的虚拟随机播放操作组合以形成一个或多个组合的虚拟随机播放操作来折叠一个或多个虚拟洗牌树,其中每个虚拟随机播放树是仅包含表达式树的子树 虚拟随机操作。 然后为一个或多个组合的虚拟随机操作生成代码。