利索-全球专利检索

  1. 登录
  2. 注册

搜索结果

8 条记录 (耗时 0.018 秒)

    Managing data accesses in storage systems
    1.
    发明授权
    Managing data accesses in storage systems 有权

    公开(公告)号：US10684944B1

    公开(公告)日：2020-06-16

    申请号：US13570827

    申请日：2012-08-09

    申请人： Khang Can ,  Qin Tao ,  Xiangping Chen ,  Siyu Zhou ,  Sadagopan Balaraman

    发明人： Khang Can ,  Qin Tao ,  Xiangping Chen ,  Siyu Zhou ,  Sadagopan Balaraman

    IPC分类号： G06F12/02 ,  G06F9/50

    摘要： A method is used in managing data accesses in storage systems. An access path to a storage object is evaluated. The storage object is associated with a first storage processor of a storage system for providing access to the storage object. The storage object is accessed using a second storage processor of the storage system. The storage system includes source and destination storage pools configured such that performance characteristics associated with the source storage pool is different from performance characteristics associated with the destination storage pool. The storage object resides in the source storage pool associated with the first storage processor. Based on the evaluation, the access path of the storage object is changed to the second storage processor based on a policy. The policy determines available storage space in the destination storage pool associated with the second storage processor for migrating the storage object to the destination storage pool.

    Scheduling data relocation activities using configurable fairness criteria
    2.
    发明授权
    Scheduling data relocation activities using configurable fairness criteria 有权
    使用可配置的公平标准调度数据迁移活动

    公开(公告)号：US08327103B1

    公开(公告)日：2012-12-04

    申请号：US12824380

    申请日：2010-06-28

    申请人： Khang Can ,  Xiangping Chen ,  Monica Chaudhary ,  Kevin Bangyen Jiang ,  Qin Tao ,  Mark Ku

    发明人： Khang Can ,  Xiangping Chen ,  Monica Chaudhary ,  Kevin Bangyen Jiang ,  Qin Tao ,  Mark Ku

    IPC分类号： G06F12/00

    CPC分类号： G06F3/061 ,  G06F3/0649 ,  G06F3/0659 ,  G06F3/0685

    摘要： A storage processor of a data storage system includes a data relocation task scheduler that provides system limit control on the execution of data relocation requests. The data relocation task scheduler also provides fairness and concurrency limit enforcement by scheduling relocation requests from different provisioning domains, such as storage pools, in a fair manner. For example, the data relocation task scheduler includes stackable sets of pending task queues that store incoming data relocation requests. The data relocation task scheduler applies an iteration scheme to remove data relocation requests from the pending task queues thereby providing scheduling fairness among the queues and among the different provisioning domains within the data storage system. The data relocation task scheduler is configured to receive updated policies that control both system-wide data relocation resource consumption and scheduling fairness among the queues during operation and dynamically adjust current policies with the updated policies.

    摘要翻译： 数据存储系统的存储处理器包括数据重定位任务调度器，其提供对数据重定位请求的执行的系统限制控制。 数据重定位任务调度器还通过以公平的方式调度来自不同供应域（例如存储池）的重定位请求来提供公平性和并发限制强制。 例如，数据重定位任务调度器包括可堆叠的待处理任务队列集合，用于存储进入的数据重定位请求。 数据重定位任务调度器应用迭代方案来从挂起的任务队列中移除数据重定位请求，从而在队列之间以及数据存储系统内的不同供应域之间提供调度公平性。 数据重定位任务调度器被配置为接收在操作期间在队列之间控制系统范围的数据重定位资源消耗和调度公平性的更新策略，并且利用更新的策略动态地调整当前策略。

    Multi-Ranker For Search
    6.
    发明申请
    Multi-Ranker For Search 有权
    搜索多域名

    公开(公告)号：US20090083248A1

    公开(公告)日：2009-03-26

    申请号：US11859066

    申请日：2007-09-21

    申请人： Tie-Yan Liu ,  Qin Tao ,  Hang Li

    发明人： Tie-Yan Liu ,  Qin Tao ,  Hang Li

    IPC分类号： G06F17/30

    CPC分类号： G06F17/3053

    摘要： Systems and methods for processing user queries and identifying a set of documents relevant to the user query from a database using multi ranker search are described. In one implementation, the retrieved documents can be paired to form document pairs, or instance pairs, in a variety of combinations. Such instance pairs may have a rank order between them as they all have different ranks. A classifier, hyperplane, and a base ranker may be constructed for identifying the rank order relationships between the two instances in an instance pair. The base ranker may be generated for each rank pair. The systems use a divide and conquer strategy for learning to rank the instance pairs by employing multiple hyperplanes and aggregate the base rankers to form an ensemble of base rankers. Such an ensemble of base rankers can be used to rank the documents or instances.

    摘要翻译： 描述了用于处理用户查询的系统和方法，以及使用多游标搜索从数据库识别与用户查询相关的一组文档。 在一个实现中，检索到的文档可以被配对以形成各种组合的文档对或实例对。 这样的实例对可以在它们之间具有排序，因为它们都具有不同的等级。 可以构造一个分类器，超平面和基本游标，用于识别实例对中的两个实例之间的排序关系。 可以为每个等级对生成基本杀手。 系统使用分裂和征服策略来学习通过使用多个超平面来对实例对进行排名，并且聚合基本等级以形成基本等级的组合。 可以使用这样一个基本排名的组合对文档或实例进行排名。

    Techniques using an encryption tier property in a multi-tiered storage environment
    7.
    发明授权
    Techniques using an encryption tier property in a multi-tiered storage environment 有权
    在多层存储环境中使用加密层属性的技术

    公开(公告)号：US09436834B1

    公开(公告)日：2016-09-06

    申请号：US13690069

    申请日：2012-11-30

    申请人： Xuan Tang ,  Qin Tao

    发明人： Xuan Tang ,  Qin Tao

    IPC分类号： G06F11/30 ,  G06F21/60

    CPC分类号： G06F21/602

    摘要： Techniques are described for storing data. A plurality of storage tiers are provided including a first set and a second set of storage tiers of physical devices. Data stored on any physical device in the first set is stored in an encrypted form. Data stored on any physical device in the second set is not stored in an encrypted form. A first value is specified for a first setting that is any of a tiering preference and tiering requirement indicating that at least one data portion of a logical device is to be stored on physical device(s) of a storage tier storing data in an encrypted form. Responsive to specifying the first value as the first setting, the at least one data portion of the logical device currently stored on physical device(s) of the second set are relocated to physical device(s) of the first set.

    摘要翻译： 描述了用于存储数据的技术。 提供了多个存储层，包括物理设备的第一组和第二组存储层。 存储在第一组中的任何物理设备上的数据以加密形式存储。 存储在第二组中的任何物理设备上的数据不以加密形式存储。 对于作为分层优先级和分层要求中的任何一个的第一设置指定第一值，指示逻辑设备的至少一个数据部分将被存储在以加密形式存储数据的存储层的物理设备上 。 响应于将第一值指定为第一设置，当前存储在第二组的物理设备上的逻辑设备的至少一个数据部分被重新定位到第一组的物理设备。

    Multi-ranker for search
    8.
    发明授权
    Multi-ranker for search 有权
    多人游戏搜索

    公开(公告)号：US08122015B2

    公开(公告)日：2012-02-21

    申请号：US11859066

    申请日：2007-09-21

    申请人： Tie-Yan Liu ,  Qin Tao ,  Hang Li

    发明人： Tie-Yan Liu ,  Qin Tao ,  Hang Li

    IPC分类号： G06F7/00 ,  G06F17/30

    CPC分类号： G06F17/3053

    摘要： Systems and methods for processing user queries and identifying a set of documents relevant to the user query from a database using multi ranker search are described. In one implementation, the retrieved documents can be paired to form document pairs, or instance pairs, in a variety of combinations. Such instance pairs may have a rank order between them as they all have different ranks. A classifier, hyperplane, and a base ranker may be constructed for identifying the rank order relationships between the two instances in an instance pair. The base ranker may be generated for each rank pair. The systems use a divide and conquer strategy for learning to rank the instance pairs by employing multiple hyperplanes and aggregate the base rankers to form an ensemble of base rankers. Such an ensemble of base rankers can be used to rank the documents or instances.

    摘要翻译： 描述了用于处理用户查询的系统和方法，以及使用多游标搜索从数据库识别与用户查询相关的一组文档。 在一个实现中，检索到的文档可以被配对以形成各种组合的文档对或实例对。 这样的实例对可以在它们之间具有排序，因为它们都具有不同的等级。 可以构造一个分类器，超平面和基本游标，用于识别实例对中的两个实例之间的排序关系。 可以为每个等级对生成基本杀手。 系统使用分裂和征服策略来学习通过使用多个超平面来对实例对进行排名，并且聚合基本等级以形成基本等级的组合。 可以使用这样一个基本排名的组合对文档或实例进行排名。