利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

18 条记录 (耗时 0.019 秒)

    Optimization of native buffer accesses in Java applications on hybrid systems
    13.
    发明授权
    Optimization of native buffer accesses in Java applications on hybrid systems 有权
    优化混合系统中Java应用程序中的本地缓冲区访问

    公开(公告)号：US09158497B2

    公开(公告)日：2015-10-13

    申请号：US13732820

    申请日：2013-01-02

    申请人： International Business Machines Corporation

    发明人： Michael H. Dawson ,  Yuqing Gao ,  Megumi Ito ,  Graeme Johnson ,  Seetharami R. Seelam

    IPC分类号： G06F12/00 ,  G06F13/00 ,  G06F13/28 ,  G06F5/08 ,  G06F3/06 ,  G06F9/54 ,  G06F11/34

    CPC分类号： G06F5/08 ,  G06F3/0685 ,  G06F9/544 ,  G06F9/545 ,  G06F11/34 ,  G06F11/3433 ,  G06F11/3452

    摘要： Managing buffers in a hybrid system, in one aspect, may comprise selecting a first buffer management method from a plurality of buffer management methods; capturing statistics associated with access to the buffer in the hybrid system running under the initial buffer management method; analyzing the captured statistics; identifying a second buffer management method based on the analyzed captured statistics; determining whether the second buffer management method is more optimal than the first buffer management method; in response to determining that the second buffer management method is more optimal than the first buffer management method, invoking the second buffer management method; and repeating the capturing, the analyzing, the identifying and the determining.

    摘要翻译： 在一个方面，在混合系统中管理缓冲器可以包括从多个缓冲器管理方法中选择第一缓冲器管理方法; 捕获在初始缓冲管理方法下运行的混合系统中对缓冲区的访问相关的统计信息; 分析捕获的统计数据; 基于分析的捕获统计识别第二缓冲管理方法; 确定所述第二缓冲器管理方法是否比所述第一缓冲器管理方法更优化; 响应于确定第二缓冲器管理方法比第一缓冲器管理方法更优化，调用第二缓冲器管理方法; 并重复捕获，分析，识别和确定。

    Scalable refresh for asymmetric non-volatile memory-based neuromorphic circuits
    14.
    发明授权
    Scalable refresh for asymmetric non-volatile memory-based neuromorphic circuits 有权

    公开(公告)号：US10740673B2

    公开(公告)日：2020-08-11

    申请号：US16539038

    申请日：2019-08-13

    申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

    发明人： Megumi Ito ,  Masatoshi Ishii ,  Atsuya Okazaki

    IPC分类号： G06N3/063 ,  G06N3/08 ,  H01L45/00 ,  H01L27/24 ,  G11C13/00

    摘要： A computer-implemented method is provided for refreshing cells in a Non-Volatile Memory (NVM)-based neuromorphic circuit wherein synapses are each composed of a respective cell pair formed from a respective Gp cell and a respective Gm cell of the cells. The method includes randomly selecting multiple neurons and reading a conductance of any of the synapses connected to the multiple neurons. The method further includes selecting any of the synapses connected to the selected multiple neurons for which the Gm conductance has reached a maximum conductance. The method also includes resetting the Gp cell and Gm cell of the selected synapses, and setting, at most, one of the Gp cell and Gm cell of each of the selected synapses to recover an effective total weight of each of the selected synapses.

    METHOD, DEVICE, AND COMPUTER PROGRAM FOR MERGE-SORTING RECORD GROUPS HAVING TREE STRUCTURE EFFICIENTLY
    17.
    发明申请
    METHOD, DEVICE, AND COMPUTER PROGRAM FOR MERGE-SORTING RECORD GROUPS HAVING TREE STRUCTURE EFFICIENTLY 有权
    方法，装置和计算机程序，用于具有有效结构的合并记录组

    公开(公告)号：US20150178338A1

    公开(公告)日：2015-06-25

    申请号：US14569062

    申请日：2014-12-12

    申请人： International Business Machines Corporation

    发明人： Hiroshige Hayashizaki ,  Megumi Ito

    IPC分类号： G06F17/30

    CPC分类号： G06F7/16 ,  G06F7/32 ,  G06F17/30961

    摘要： A first record group and a second record group having a tree structure are merge-sorted. The first record group and the second record group are acquired, and depth information indicative of the hierarchical depth of the tree structure is attached as metadata to key information on each node of each record included in the acquired first record group and second record group. The depth information is compared in preference to the key information to perform merge sort sequentially.

    摘要翻译： 具有树结构的第一个记录组和第二个记录组被合并排序。 获取第一记录组和第二记录组，并且指示树结构的层次深度的深度信息作为元数据附加到包含在所获取的第一记录组和第二记录组中的每个记录的每个记录的关键信息。 将深度信息与优先顺序地进行合并排序的密钥信息进行比较。

    JAVA NATIVE INTERFACE ARRAY HANDLING IN A DISTRIBUTED JAVA VIRTUAL MACHINE
    18.
    发明申请
    JAVA NATIVE INTERFACE ARRAY HANDLING IN A DISTRIBUTED JAVA VIRTUAL MACHINE 有权

    公开(公告)号：US20140068579A1

    公开(公告)日：2014-03-06

    申请号：US13844241

    申请日：2013-03-15

    申请人： INTERNATIONAL BUSINESS MACHINES CORPORATION

    发明人： Michael Hilton Dawson ,  Xavier Rene Guerin ,  Megumi Ito ,  Graeme Johnson ,  Seetharami R. Seelam

    IPC分类号： G06F9/455

    CPC分类号： G06F9/45504 ,  G06F9/547

    摘要： A method for executing native code in a distributed Java Virtual Machine (JVM) is disclosed herein. The method may include receiving, in a first thread executing in a remote execution container, a first native code-generated call, such as a Java Native Interface (JNI) call, to a second thread, the first call including a first array write request. The first call may be stored in an instruction cache and bundled with a second native code-generated call and sent to the second thread. The calls are unbundled and executed in the second thread. An opaque handle to an array returned by the second call is bundled with corresponding array data and returned to the first thread. The array data of the bundle is stored in a data cache and retrieved in response to requests for the array data addressed to the second thread. A corresponding computer program product is also disclosed.