公开(公告)号：US09164806B2

公开(公告)日：2015-10-20

申请号：US13359391

申请日：2012-01-26

申请人： Brian Oliver ,  Christer Fahlgren ,  Noah Arliss

发明人： Brian Oliver ,  Christer Fahlgren ,  Noah Arliss

IPC分类号： G06F15/173 ,  G06F9/50

CPC分类号： G06F9/5038 ,  G06F9/5027

摘要： A processing pattern is described for dispatching and executing tasks in a distributed computing grid, such as a cluster network. The grid includes a plurality of computer nodes that store a set of data and perform operations on that data. The grid provides an interface that allows clients to submit tasks to the cluster for processing. The interface can be used to establish a session between the client and the cluster, which will be used to submit a task for processing by the plurality of computer nodes of the cluster. A dispatcher receives a submission of the task over the interface and routes the task to at least one node in the cluster that is designated to process the task. A task processor then processes the task on the designated node(s), generates a submission outcome and indicates to the client that the submission outcome is available.

摘要翻译： 描述了用于在诸如群集网络的分布式计算网格中调度和执行任务的处理模式。 网格包括存储一组数据并对该数据执行操作的多个计算机节点。 网格提供了一个接口，允许客户端将任务提交到集群进行处理。 该接口可用于在客户端和集群之间建立会话，该会话将用于提交由集群的多个计算机节点处理的任务。 调度员通过接口接收到任务的提交，并将任务路由到集群中至少一个指定为处理任务的节点。 然后，任务处理器处理指定节点上的任务，生成提交结果，并向客户端指示提交结果可用。

公开(公告)号：US20120197959A1

公开(公告)日：2012-08-02

申请号：US13359391

申请日：2012-01-26

申请人： Brian Oliver ,  Christer Fahlgren ,  Noah Arliss

发明人： Brian Oliver ,  Christer Fahlgren ,  Noah Arliss

IPC分类号： G06F15/16 ,  G06F9/48

CPC分类号： G06F9/5038 ,  G06F9/5027

摘要： A processing pattern is described for dispatching and executing tasks in a distributed computing grid, such as a cluster network. The grid includes a plurality of computer nodes that store a set of data and perform operations on that data. The grid provides an interface that allows clients to submit tasks to the cluster for processing. The interface can be used to establish a session between the client and the cluster, which will be used to submit a task for processing by the plurality of computer nodes of the cluster. A dispatcher receives a submission of the task over the interface and routes the task to at least one node in the cluster that is designated to process the task. A task processor then processes the task on the designated node(s), generates a submission outcome and indicates to the client that the submission outcome is available.

摘要翻译： 描述了用于在诸如群集网络的分布式计算网格中调度和执行任务的处理模式。 网格包括存储一组数据并对该数据执行操作的多个计算机节点。 网格提供了一个接口，允许客户端将任务提交到集群进行处理。 该接口可用于在客户端和集群之间建立会话，该会话将用于提交由集群的多个计算机节点处理的任务。 调度员通过接口接收到任务的提交，并将任务路由到集群中至少一个指定为处理任务的节点。 然后，任务处理器处理指定节点上的任务，生成提交结果，并向客户端指示提交结果可用。

公开(公告)号：US09621409B2

公开(公告)日：2017-04-11

申请号：US13462720

申请日：2012-05-02

申请人： Brian Oliver ,  Noah Arliss ,  Gene Gleyzer ,  Robert H. Lee

发明人： Brian Oliver ,  Noah Arliss ,  Gene Gleyzer ,  Robert H. Lee

IPC分类号： G06F17/30 ,  H04L12/24 ,  G06F9/54 ,  H04L29/08 ,  H04L29/06

CPC分类号： H04L41/0631 ,  G06F9/542 ,  G06F17/30575 ,  G06F2209/542 ,  H04L67/1097 ,  H04L67/16 ,  H04L67/42

摘要： A system and method can handle storage events in a distributed data grid. The distributed data grid cluster includes a plurality of cluster nodes storing data partitions distributed throughout the cluster, each cluster node being responsible for a set of partitions. A service thread, executing on at least one of said cluster nodes in the distributed data grid, is responsible for handling one or more storage events. The service thread can use a worker thread to accomplish synchronous event handling without blocking the service thread.

公开(公告)号：US09348668B2

公开(公告)日：2016-05-24

申请号：US13462719

申请日：2012-05-02

申请人： Brian Oliver ,  Noah Arliss ,  Gene Gleyzer ,  Robert H. Lee

发明人： Brian Oliver ,  Noah Arliss ,  Gene Gleyzer ,  Robert H. Lee

IPC分类号： G06F9/46 ,  G06F9/54

CPC分类号： H04L41/0631 ,  G06F9/542 ,  G06F17/30575 ,  G06F2209/542 ,  H04L67/1097 ,  H04L67/16 ,  H04L67/42

摘要： A server-side event model provides a general-purpose event framework which simplifies the server-side programming model in a distributed data grid storing data partitions distributed throughout a cluster of nodes. A system provides event interceptors which handle events associated with operations and maps the event interceptors to event dispatchers placed in the cluster. Advantageously, the system supports handling critical path events without the need for interactions from the client-side thereby avoiding unnecessary delays waiting for client responses. Additionally, the system can defer completion of an operation in the distributed data grid pending completion of event handling by an event interceptor. The system enables the data grid to employ more types of events and define different event interceptors for handling the events while avoiding client interaction overhead.

摘要翻译： 服务器端事件模型提供了一个通用事件框架，它简化了分布式数据网格中的服务器端编程模型，从而存储分布在整个节点集群中的数据分区。 系统提供事件拦截器，其处理与操作相关联的事件，并将事件拦截器映射到放置在群集中的事件分派器。 有利地，系统支持处理关键路径事件，而不需要来自客户端的交互，从而避免等待客户端响应的不必要的延迟。 此外，系统可以延迟完成分布式数据网格中的操作，等待事件拦截器完成事件处理。 该系统使数据网格能够采用更多类型的事件，并定义不同的事件拦截器来处理事件，同时避免客户端的交互开销。

公开(公告)号：US09081839B2

公开(公告)日：2015-07-14

申请号：US13360487

申请日：2012-01-27

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F7/00 ,  G06F17/30

CPC分类号： G06F17/30575

摘要： Push replication techniques are described for use in an in-memory data grid. When applications on a cluster perform insert, update or delete operations in the cache, a push replication provider asynchronously pushes updates from the source cluster to one or more remote destination clusters. The push replication provider includes a pluggable internal transport to send the updates to the destination cluster. This pluggable transport can be switched to employ a different communication service or protocol. A publishing transformer can chain multiple filters and apply filters on a stream of updates from source cluster to the destination cluster. A batch publisher can be used to receive batches multiple updates and replicate those batch to the destination cluster. XML based configuration can be provided to configure the push replication techniques on a cluster. A number of cluster topologies can be utilized, including active/passive, active/active, multi-site active/passive, multi-site active/active and centralized replication arrangement.

摘要翻译： 描述推送复制技术用于内存数据网格。 当集群上的应用程序在高速缓存中执行插入，更新或删除操作时，推送复制提供程序将异步地将更新从源集群推送到一个或多个远程目标集群。 推送复制提供程序包含一个可插拔的内部传输，用于将更新发送到目标群集。 该可插拔传输可以被切换以采用不同的通信服务或协议。 出版变压器可以链接多个过滤器，并在源集群到目标集群的更新流上应用过滤器。 可以使用批处理发布者接收批次多个更新，并将这些批复制到目标集群。 可以提供基于XML的配置来配置群集上的推送复制技术。 可以使用许多集群拓扑，包括主动/被动，主动/主动，多站点主动/被动，多站点主动/主动和集中复制安排。

公开(公告)号：US20120278817A1

公开(公告)日：2012-11-01

申请号：US13359396

申请日：2012-01-26

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F9/54

CPC分类号： G06F9/4843

摘要： An event distribution pattern is described for use with a distributed data grid. The grid can be comprised of a cluster of computer devices having a cache for storing data entries. An event distributor residing on at least one of those computer devices provides a domain for sending events to a desired end point destination and also provides the store and forward semantics for ensuring asynchronous delivery of those events. An event channel controller resides as an entry in the cache on at least one of computers in the cluster. This event channel controller receives the events defined by said application from the event distributor and provides the events to a set of event channels. Each event channel controller can include multiple event channel implementations for distributing the events to different destinations. The destinations can include local caches, remote caches, standard streams, files and JMS components

摘要翻译： 描述了与分布式数据网格一起使用的事件分发模式。 网格可以由具有用于存储数据条目的高速缓存的计算机设备的集群组成。 驻留在这些计算机设备中的至少一个的事件分发器提供用于将事件发送到期望的终点目的地的域，并且还提供存储和转发语义以确保这些事件的异步递送。 事件通道控制器驻留在集群中的至少一台计算机上的高速缓存中的条目。 该事件通道控制器从事件分发器接收由所述应用程序定义的事件，并将事件提供给一组事件通道。 每个事件通道控制器可以包括用于将事件分发到不同目的地的多个事件通道实现。 目的地可以包括本地缓存，远程缓存，标准流，文件和JMS组件

公开(公告)号：US20120278422A1

公开(公告)日：2012-11-01

申请号：US13359399

申请日：2012-01-26

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F15/167

CPC分类号： G06F9/4843

摘要： A live object pattern is described that enables a distributed cache to store live objects as data entries thereon. A live object is a data entry stored in the distributed cache which represents a particular function or responsibility. When a live object arrives to the cache on a particular cluster server, a set of interfaces are called back which inform the live object that it has arrived at that server and that it should initiate to perform its functions. A live object is thus different from “dead” data entries because a live object performs a set of function, can be started/stopped and can interact with other live objects in the distributed cache. Because live objects are backed up across the cluster just like normal data entries, the functional components of the system are more highly available and are easily transferred to another server's cache in case of failures.

摘要翻译： 描述了一种活动对象模式，其使分布式高速缓存能够将活动对象存储为数据条目。 活动对象是存储在分布式缓存中的数据条目，其表示特定的功能或责任。 当活动对象到达特定群集服务器上的缓存时，将回调一组接口，该对接通知活动对象已到达该服务器，并且该对象应启动执行其功能。 因此，活动对象与死亡数据条目不同，因为活动对象执行一组功能，可以启动/停止并且可以与分布式缓存中的其他活动对象进行交互。 由于活动对象与集群中的常规数据条目一样进行备份，所以系统的功能组件更具可用性，并且在发生故障时可轻松转移到其他服务器的缓存。

公开(公告)号：US10127077B2

公开(公告)日：2018-11-13

申请号：US13359396

申请日：2012-01-26

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F9/48

摘要： An event distribution pattern is described for use with a distributed data grid. The grid can be comprised of a cluster of computer devices having a cache for storing data entries. An event distributor residing on at least one of those computer devices provides a domain for sending events to a desired end point destination and also provides the store and forward semantics for ensuring asynchronous delivery of those events. An event channel controller resides as an entry in the cache on at least one of computers in the cluster. This event channel controller receives the events defined by said application from the event distributor and provides the events to a set of event channels. Each event channel controller can include multiple event channel implementations for distributing the events to different destinations. The destinations can include local caches, remote caches, standard streams, files and JMS components.

公开(公告)号：US20120197840A1

公开(公告)日：2012-08-02

申请号：US13360487

申请日：2012-01-27

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F7/00

CPC分类号： G06F17/30575

摘要： Push replication techniques are described for use in an in-memory data grid. When applications on a cluster perform insert, update or delete operations in the cache, a push replication provider asynchronously pushes updates from the source cluster to one or more remote destination clusters. The push replication provider includes a pluggable internal transport to send the updates to the destination cluster. This pluggable transport can be switched to employ a different communication service or protocol. A publishing transformer can chain multiple filters and apply filters on a stream of updates from source cluster to the destination cluster. A batch publisher can be used to receive batches multiple updates and replicate those batch to the destination cluster. XML based configuration can be provided to configure the push replication techniques on a cluster. A number of cluster topologies can be utilized, including active/passive, active/active, multi-site active/passive, multi-site active/active and centralized replication arrangement.

摘要翻译： 描述推送复制技术用于内存数据网格。 当集群上的应用程序在高速缓存中执行插入，更新或删除操作时，推送复制提供程序将异步地将更新从源集群推送到一个或多个远程目标集群。 推送复制提供程序包含一个可插拔的内部传输，用于将更新发送到目标群集。 该可插拔传输可以被切换以采用不同的通信服务或协议。 出版变压器可以链接多个过滤器，并在源集群到目标集群的更新流上应用过滤器。 可以使用批处理发布者接收批次多个更新，并将这些批复制到目标集群。 可以提供基于XML的配置来配置群集上的推送复制技术。 可以使用许多集群拓扑，包括主动/被动，主动/主动，多站点主动/被动，多站点主动/主动和集中复制安排。

公开(公告)号：US08843581B2

公开(公告)日：2014-09-23

申请号：US13359399

申请日：2012-01-26

申请人： Brian Oliver ,  Noah Arliss

发明人： Brian Oliver ,  Noah Arliss

IPC分类号： G06F15/167 ,  G06F9/48

CPC分类号： G06F9/4843

摘要： A live object pattern is described that enables a distributed cache to store live objects as data entries thereon. A live object is a data entry stored in the distributed cache which represents a particular function or responsibility. When a live object arrives to the cache on a particular cluster server, a set of interfaces are called back which inform the live object that it has arrived at that server and that it should initiate to perform its functions. A live object is thus different from “dead” data entries because a live object performs a set of function, can be started/stopped and can interact with other live objects in the distributed cache. Because live objects are backed up across the cluster just like normal data entries, the functional components of the system are more highly available and are easily transferred to another server's cache in case of failures.

摘要翻译： 描述了一种活动对象模式，其使分布式高速缓存能够将活动对象存储为数据条目。 活动对象是存储在分布式缓存中的数据条目，其表示特定的功能或责任。 当活动对象到达特定群集服务器上的缓存时，将回调一组接口，该对接通知活动对象已到达该服务器，并且该对象应启动执行其功能。 因此，活动对象与“死”数据条目不同，因为活动对象执行一组功能，可以启动/停止，并可以与分布式缓存中的其他活动对象进行交互。 由于活动对象与集群中的常规数据条目一样进行备份，所以系统的功能组件更具可用性，并且在发生故障时可轻松转移到其他服务器的缓存。