Mechanism for preserving session state when using an access-limited buffer
    3.
    发明申请
    Mechanism for preserving session state when using an access-limited buffer 有权
    使用访问限制缓冲区时保留会话状态的机制

    公开(公告)号:US20060174011A1

    公开(公告)日:2006-08-03

    申请号:US11049071

    申请日:2005-02-01

    IPC分类号: G06F15/16

    摘要: Preserving session state correlation in a distributed computing environment in which a service module instance receives messages from a general input buffer. The general input buffer is not accessed by a service module instance one the service module instance is determined to be recycled. After the service module instance receives a session initiation message through the general input buffer, a session-oriented input buffer is associated with the service module instance. Any subsequent session messages are then received by the service module via the session-oriented input buffer. Even if the service module instance was determined to be recycled mid-session, the service module instance would still operate to process all of the remaining session messages from the session input buffer, thereby preserving session continuity.

    摘要翻译: 在服务模块实例从一般输入缓冲区接收消息的分布式计算环境中保留会话状态相关性。 通常的输入缓冲区不被服务模块实例访问,服务模块实例被确定为被回收。 在服务模块实例通过通用输入缓冲器接收到会话发起消息之后,面向会话的输入缓冲器与服务模块实例相关联。 随后的会话消息随后由服务模块经由面向会话的输入缓冲器接收。 即使服务模块实例被确定为在会话中继续使用,服务模块实例仍然将操作以从会话输入缓冲器处理所有剩余的会话消息,从而保持会话连续性。

    Mechanism for preserving session state when using an access-limited buffer
    4.
    发明授权
    Mechanism for preserving session state when using an access-limited buffer 有权
    使用访问限制缓冲区时保留会话状态的机制

    公开(公告)号:US07565395B2

    公开(公告)日:2009-07-21

    申请号:US11049071

    申请日:2005-02-01

    IPC分类号: G06F15/16 G06F15/173

    摘要: Preserving session state correlation in a distributed computing environment in which a service module instance receives messages from a general input buffer. The general input buffer is not accessed by a service module instance once the service module instance is determined to be recycled. After the service module instance receives a session initiation message through the general input buffer, a session-oriented input buffer is associated with the service module instance. Any subsequent session messages are then received by the service module via the session-oriented input buffer. Even if the service module instance was determined to be recycled mid-session, the service module instance would still operate to process all of the remaining session messages from the session input buffer, thereby preserving session continuity.

    摘要翻译: 在服务模块实例从一般输入缓冲区接收消息的分布式计算环境中保留会话状态相关性。 一旦确定服务模块实例被回收,服务模块实例就不会访问通用输入缓冲区。 在服务模块实例通过通用输入缓冲器接收到会话发起消息之后,面向会话的输入缓冲器与服务模块实例相关联。 随后的会话消息随后由服务模块经由面向会话的输入缓冲器接收。 即使服务模块实例被确定为在会话中继续使用,服务模块实例仍然将操作以从会话输入缓冲器处理所有剩余的会话消息,从而保持会话连续性。

    Software agent-based architecture for data relocation
    5.
    发明申请
    Software agent-based architecture for data relocation 有权
    用于数据迁移的基于软件代理的架构

    公开(公告)号:US20070083725A1

    公开(公告)日:2007-04-12

    申请号:US11244702

    申请日:2005-10-06

    IPC分类号: G06F13/00

    摘要: Data is relocated among data storage nodes in a data center by using distributed agents. An agent at a first data storage node manages the relocation of associated data to a second data storage node, as well as the relocation of a mirror copy of the data at a third data storage node to a fourth data storage node, by receiving status information, such as via a publish-subscribe technique. The first agent may provide an identifier with a relocation request to the third data storage node which is relayed to the fourth data storage node, and used in status messages provided by the third and fourth data storage nodes. The received status information is then associated with the relocation request. Agents for each pair of sending and receiving data storage nodes can monitor their availability for servicing the relocation to negotiate an acceptable transmission rate.

    摘要翻译: 通过使用分布式代理将数据重新定位在数据中心的数据存储节点之间。 第一数据存储节点处的代理管理关联数据到第二数据存储节点的重定位,以及通过接收状态信息将第三数据存储节点处的数据的镜像副本重定位到第四数据存储节点 ,例如通过发布订阅技术。 第一代理可以向中继到第四数据存储节点的第三数据存储节点提供重定位请求的标识符,并且用于由第三和第四数据存储节点提供的状态消息中。 接收到的状态信息然后与重定位请求相关联。 每对发送和接收数据存储节点的代理可以监视其可用性以服务重定位以协商可接受的传输速率。

    Peer-to-peer auction based data distribution
    6.
    发明申请
    Peer-to-peer auction based data distribution 有权
    基于点对点拍卖的数据分发

    公开(公告)号:US20070088703A1

    公开(公告)日:2007-04-19

    申请号:US11252002

    申请日:2005-10-17

    IPC分类号: G06F17/30

    摘要: A peer-to-peer auction based strategy balances the storage of data in a data center. Each cluster of the data center has a cluster manager and associated data storage nodes on which data is stored. The data storage nodes can periodically announce their respective utilization levels to other data storage nodes. If the announcing nodes are over capacity, the other nodes in the cluster respond by checking their utilization level and placing a bid if they are able to accept new data. Data is then relocated to one or more selected bidding nodes. If data cannot be balanced at the data storage node level, the cluster manager can identify nodes in other clusters that are able to accept new data. Multiple auctions for the same or different categories of data can be conducted at the same time by the data storage nodes and/or cluster managers.

    摘要翻译: 基于点对点竞价的策略平衡数据中心数据的存储。 数据中心的每个集群都有一个集群管理器和数据存储在其上的关联数据存储节点。 数据存储节点可以周期性地向其他数据存储节点通告其各自的利用水平。 如果通告节点超过容量,集群中的其他节点将通过检查其使用级别进行响应,并在能够接受新数据时进行投标。 然后将数据重新定位到一个或多个选定的出价节点。 如果数据在数据存储节点级别无法平衡,则集群管理器可以识别能够接受新数据的其他集群中的节点。 相同或不同类别的数据的多次拍卖可以由数据存储节点和/或集群管理器同时进行。

    Managing requests based on request groups
    7.
    发明授权
    Managing requests based on request groups 有权
    根据请求组管理请求

    公开(公告)号:US08726284B2

    公开(公告)日:2014-05-13

    申请号:US12813382

    申请日:2010-06-10

    IPC分类号: G06F9/46

    CPC分类号: G06F9/5027

    摘要: A request management component receives requests to perform an operation. Each of the requests is assigned, based on one or more criteria, to one of multiple different request groups. Based at least in part on execution policies associated with the request groups, determinations are made as to when to submit the requests to one or more recipient. Each of the multiple requests is submitted to one of the recipients when it is determined that the request is to be submitted.

    摘要翻译: 请求管理组件接收执行操作的请求。 根据一个或多个标准将每个请求分配给多个不同的请求组之一。 至少部分地基于与请求组相关联的执行策略,确定何时将请求提交给一个或多个接收者。 当确定要提交请求时,将多个请求中的每一个提交给其中一个收件人。

    Managing Requests Based on Request Groups
    9.
    发明申请
    Managing Requests Based on Request Groups 有权
    基于请求组管理请求

    公开(公告)号:US20110307895A1

    公开(公告)日:2011-12-15

    申请号:US12813382

    申请日:2010-06-10

    IPC分类号: G06F9/46

    CPC分类号: G06F9/5027

    摘要: A request management component receives requests to perform an operation. Each of the requests is assigned, based on one or more criteria, to one of multiple different request groups. Based at least in part on execution policies associated with the request groups, determinations are made as to when to submit the requests to one or more recipient. Each of the multiple requests is submitted to one of the recipients when it is determined that the request is to be submitted.

    摘要翻译: 请求管理组件接收执行操作的请求。 根据一个或多个标准将每个请求分配给多个不同的请求组之一。 至少部分地基于与请求组相关联的执行策略,确定何时将请求提交给一个或多个接收者。 当确定要提交请求时,将多个请求中的每一个提交给其中一个收件人。

    Peer-to-peer auction based data distribution
    10.
    发明授权
    Peer-to-peer auction based data distribution 有权
    基于点对点拍卖的数据分发

    公开(公告)号:US07558859B2

    公开(公告)日:2009-07-07

    申请号:US11252002

    申请日:2005-10-17

    IPC分类号: G06F15/173

    摘要: A peer-to-peer auction based strategy balances the storage of data in a data center. Each cluster of the data center has a cluster manager and associated data storage nodes on which data is stored. The data storage nodes can periodically announce their respective utilization levels to other data storage nodes. If the announcing nodes are over capacity, the other nodes in the cluster respond by checking their utilization level and placing a bid if they are able to accept new data. Data is then relocated to one or more selected bidding nodes. If data cannot be balanced at the data storage node level, the cluster manager can identify nodes in other clusters that are able to accept new data. Multiple auctions for the same or different categories of data can be conducted at the same time by the data storage nodes and/or cluster managers.

    摘要翻译: 基于点对点竞价的策略平衡数据中心数据的存储。 数据中心的每个集群都有一个集群管理器和数据存储在其上的关联数据存储节点。 数据存储节点可以周期性地向其他数据存储节点通告其各自的利用水平。 如果通告节点超过容量,集群中的其他节点将通过检查其利用水平进行响应,并在能够接受新数据时进行投标。 然后将数据重新定位到一个或多个选定的出价节点。 如果数据在数据存储节点级别无法平衡,则集群管理器可以识别能够接受新数据的其他集群中的节点。 相同或不同类别的数据的多次拍卖可以由数据存储节点和/或集群管理器同时进行。