中科微点-全球专利检索

  1. Login
  2. Sign Up

Search Results

105 records (took 0.024 seconds)

    ACTIVE MEMORY EXPANSION IN A DATABASE ENVIRONMENT TO QUERY NEEDED/UNNEEDED RESULTS
    2.
    发明申请
    ACTIVE MEMORY EXPANSION IN A DATABASE ENVIRONMENT TO QUERY NEEDED/UNNEEDED RESULTS 审中-公开
    数据库环境中的主动内存扩展查询需求/未被使用的结果

    公开(公告)号：WO2012062597A1

    公开(公告)日：2012-05-18

    申请号：PCT/EP2011/068871

    申请日：2011-10-27

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  SANTOSUOSSO, John ,  BARSNESS, Eric, Lawrence ,  NELSESTUEN, Paul, Stuart

    Inventor： SANTOSUOSSO, John ,  BARSNESS, Eric, Lawrence ,  NELSESTUEN, Paul, Stuart

    IPC: G06F17/30

    CPC classification number: G06F17/30153 ,  G06F17/30442

    Abstract: Techniques are described for estimating and managing memory compression for request processing. Embodiments of the invention may generally include receiving a request for data, determining if the requested data contains any compressed data, and sending the requesting entity only the uncompressed data. A separate embodiment generally includes receiving a request for data, determining if the requested data contains any compressed data, gathering uncompression criteria about the requested data, and using the uncompression criteria to selectively determine what portion of the compressed data to uncompress.

    Abstract translation: 描述了用于估计和管理用于请求处理的存储器压缩的技术。 本发明的实施例通常可以包括接收对数据的请求，确定所请求的数据是否包含任何压缩数据，以及仅向请求实体发送未压缩的数据。 单独的实施例通常包括接收对数据的请求，确定所请求的数据是否包含任何压缩数据，收集关于所请求数据的解压缩标准，以及使用解压缩标准来选择性地确定压缩数据的哪一部分以解压缩。

    SOFT FAILURE DETECTION
    3.
    发明申请
    SOFT FAILURE DETECTION 审中-公开
    软故障检测

    公开(公告)号：WO2012049014A1

    公开(公告)日：2012-04-19

    申请号：PCT/EP2011/066723

    申请日：2011-09-27

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  CAFFREY, James

    Inventor： CAFFREY, James

    IPC: G06F11/00 ,  G06F11/07

    CPC classification number: G06F11/076 ,  G06F11/0709 ,  G06F11/0715 ,  G06F11/0751

    Abstract: A method, system, and computer program product detect soft failures as follows. A set of artifacts being generated by at least one process in a system is monitored. A number of artifacts being generated by the process is determined to be below a given threshold in response to the monitoring. The process is monitored in response to the determination. A current state of the process is determined in response to the analyzing. A notification is generated in response to the current state of the process including a set of abnormal behaviors.

    Abstract translation: 一种方法，系统和计算机程序产品检测软故障如下。 监视由系统中的至少一个进程生成的一组工件。 响应于该监视，由该过程产生的许多伪影被确定为低于给定阈值。 响应于该确定来监视该过程。 响应于分析确定该过程的当前状态。 响应于包括一组异常行为的处理的当前状态产生通知。

    MULTI-PRIORITY COMMAND PROCESSING AMONG BACK-END PROCESSORS
    5.
    发明申请
    MULTI-PRIORITY COMMAND PROCESSING AMONG BACK-END PROCESSORS 审中-公开
    后端处理器的多重优先级命令处理

    公开(公告)号：WO2011160972A1

    公开(公告)日：2011-12-29

    申请号：PCT/EP2011/059754

    申请日：2011-06-13

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  IBM UNITED KINGDOM LIMITED ,  LONG, Thomas, Charles ,  MAKOWICKI, Robert, Philip

    Inventor： LONG, Thomas, Charles ,  MAKOWICKI, Robert, Philip

    IPC: G06F9/48

    CPC classification number: G06F9/4843

    Abstract: A method, system and computer program product for serially transmitting processor commands of different execution priority. A front-end processor, for example, serially receives processor commands. A low-priority queue coupled to the front-end processor stores low-priority commands, and a high-priority queue coupled to the front-end processor stores high-priority commands. A controller enables transmission of commands from either the low-priority queue or the high-priority queue for execution.

    Abstract translation: 一种用于串行发送具有不同执行优先级的处理器命令的方法，系统和计算机程序产品。 例如，前端处理器串行地接收处理器命令。 耦合到前端处理器的低优先级队列存储低优先级命令，并且耦合到前端处理器的高优先级队列存储高优先级命令。 控制器能够从低优先级队列或高优先级队列传输命令以执行。

    DIAGNOSTIC DATA CAPTURE IN A COMPUTING ENVIRONMENT
    8.
    发明申请
    DIAGNOSTIC DATA CAPTURE IN A COMPUTING ENVIRONMENT 审中-公开
    计算环境中的诊断数据捕获

    公开(公告)号：WO2011113828A1

    公开(公告)日：2011-09-22

    申请号：PCT/EP2011/053886

    申请日：2011-03-15

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  WRIGHT, Andrew ,  LEWIS, Roger, Gordon ,  ADDISON, Edward, Alan ,  LEE, Philip, Robert

    Inventor： WRIGHT, Andrew ,  LEWIS, Roger, Gordon ,  ADDISON, Edward, Alan ,  LEE, Philip, Robert

    IPC: G06F11/30 ,  G06F11/07

    CPC classification number: G06F11/0781 ,  G06F11/0715 ,  G06F11/0754 ,  G06F11/3055

    Abstract: A method and apparatus in a multithreaded computing environment for capturing diagnostic data, comprising: means responsive to a monitor determining that the computing environment is in a predetermined invalid state for initiating a trap delay component having a delay threshold; and means for recording, by a first thread, diagnostic data for the computing environment; wherein the delay setting delays operation of said means for recording diagnostic data until said monitor determines that said predetermined invalid state has repeated said determining with total count or for a period corresponding to a predetermined indicator that it is a state other than a transient state of the computing environment, the transient state corresponding to a transition of the computing environment by one or more other threads from a first valid state to a second valid state.

    Abstract translation: 一种用于捕获诊断数据的多线程计算环境中的方法和装置，包括：响应于监视器确定所述计算环境处于预定无效状态以启动具有延迟阈值的陷阱延迟分量的装置; 以及用于由第一线程记录所述计算环境的诊断数据的装置; 其中所述延迟设置延迟所述用于记录诊断数据的装置的操作，直到所述监视器确定所述预定无效状态已经以总计数重复所述确定，或者对于与预定指示符相对应的周期， 计算环境，对应于一个或多个其他线程从第一有效状态到第二有效状态的计算环境的转换的暂态。

    TRANSACTIONAL UPDATING IN DYNAMIC DISTRIBUTED WORKLOADS
    10.
    发明申请
    TRANSACTIONAL UPDATING IN DYNAMIC DISTRIBUTED WORKLOADS 审中-公开

    公开(公告)号：WO2011083056A3

    公开(公告)日：2011-07-14

    申请号：PCT/EP2010/070631

    申请日：2010-12-23

    Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION ,  JOHNSON, Paul ,  BURGESS, George ,  MITCHELL, Ian

    Inventor： JOHNSON, Paul ,  BURGESS, George ,  MITCHELL, Ian

    IPC: G06F9/46 ,  G06F9/50

    Abstract: A workload manager is operable with a distributed transaction processor having a plurality of processing regions (110, 120) and comprises: a transaction initiator region (100) for initiating a transaction; a transaction router component (170) for routing an initiated transaction to one of the plurality of processing regions; an affinity controller component for restricting transaction routing operations to maintain affinities; the affinity controller component characterised in comprising a unit of work affinity component operable with a resource manager at the one of the plurality of processing regions to activate an affinity responsive to completion of a recoverable data operation at the one of the plurality of processing regions.

Patent Agency Ranking