System and method for managing data in a distributed system
    2.
    发明授权
    System and method for managing data in a distributed system 有权
    用于在分布式系统中管理数据的系统和方法

    公开(公告)号:US07058773B1

    公开(公告)日:2006-06-06

    申请号:US10346352

    申请日:2003-01-16

    IPC分类号: G06F12/16

    CPC分类号: G06F11/1666

    摘要: A memory system includes a client device and multiple memory devices. The client device communicates an operation request pertaining to a particular data object to one memory device. Referencing a host table, the memory device identifies multiple memory devices associated with an object key of the data object. The memory device communicates the operation request to the identified memory devices to be completed. If one of the memory devices in the memory system becomes inaccessible, the memory system is capable of replacing the inaccessible memory device in the host table. Furthermore, if a new memory device is added to the memory system, the memory system is capable of modifying the host table to associate some of the possible object keys with the new memory device.

    摘要翻译: 存储器系统包括客户端设备和多个存储设备。 客户端设备将与特定数据对象相关的操作请求传送到一个存储设备。 引用主机表,内存设备识别与数据对象的对象密钥相关联的多个内存设备。 存储器装置将操作请求传送到要完成的所识别的存储器件。 如果存储器系统中的一个存储器件变得不可访问,则存储器系统能够替换主机表中的不可访问的存储器件。 此外,如果将新的存储器设备添加到存储器系统中,则存储器系统能够修改主机表以将一些可能的对象密钥与新的存储器设备相关联。

    System and method for managing network services in a distributed system
    4.
    发明授权
    System and method for managing network services in a distributed system 有权
    在分布式系统中管理网络服务的系统和方法

    公开(公告)号:US07860957B1

    公开(公告)日:2010-12-28

    申请号:US10326424

    申请日:2002-12-20

    IPC分类号: G06F15/173

    摘要: A system includes a network and multiple network devices. The network devices run instances of various services. If a particular network device becomes inaccessible, the system is capable of replacing the services instances running on the inaccessible network device by initiating new service instances on the remaining network devices. If a new network device becomes accessible, the network is capable of redistributing the currently running service instances to balance the workload by initiating service instances on the new network device and terminating service instances on existing network device.

    摘要翻译: 系统包括网络和多个网络设备。 网络设备运行各种服务的实例。 如果特定网络设备变得无法访问,则系统能够通过在其余网络设备上发起新的服务实例来替换在不可访问的网络设备上运行的服务实例。 如果新的网络设备可以访问,则网络能够通过在新的网络设备上启动服务实例并终止现有网络设备上的服务实例来重新分配当前运行的服务实例来平衡工作负载。

    System and method for providing redundant data load sharing in a distributed network
    5.
    发明授权
    System and method for providing redundant data load sharing in a distributed network 有权
    在分布式网络中提供冗余数据负载共享的系统和方法

    公开(公告)号:US07657781B1

    公开(公告)日:2010-02-02

    申请号:US10627858

    申请日:2003-07-25

    IPC分类号: G06F11/00

    摘要: In a distributed network, a data entry is received. The data entry is stored in a first one of a plurality of nodes and replicated in a second one of the plurality of nodes in order to provide data redundancy in the distributed network. In the event of a failure of the second one of the plurality of nodes, the data entry that was replicated therein is re-replicated into a third one of the plurality of nodes in order to maintain data redundancy despite the failure of the second one of the plurality of nodes. Data redundancy can be retained despite the failure of the second one of the plurality of nodes as long as the overall capacity for the distributed network remains above an initial threshold.

    摘要翻译: 在分布式网络中,接收数据输入。 数据条目存储在多个节点中的第一个节点中,并复制在多个节点的第二节点中,以便在分布式网络中提供数据冗余。 在多个节点中的第二个节点发生故障的情况下,将其中复制的数据条目重新复制到多个节点中的第三个节点中,以便维持数据冗余,尽管第二个 多个节点。 只要分布式网络的总容量保持高于初始阈值,则可以保留多个节点中的第二个节点的故障的数据冗余。