公开(公告)号：US07831731B2

公开(公告)日：2010-11-09

申请号：US09880631

申请日：2001-06-12

申请人： Wenting Tang ,  Ludmila Cherkasova ,  Lance Warren Russell

发明人： Wenting Tang ,  Ludmila Cherkasova ,  Lance Warren Russell

IPC分类号： G06F15/16 ,  G06F15/177 ,  G06F15/173

CPC分类号： H04L29/06 ,  H04L67/1002 ,  H04L67/1014 ,  H04L67/1017 ,  H04L69/16 ,  H04L69/161 ,  H04L69/163

摘要： A method and system for handing-off TCP states in a communication network. Specifically, the present invention allows for handing-off TCP states between nodes in an associated network that is optimized for rare handoff of TCP states. The handoff occurs between dynamically loadable modules that wrap around the TCP/IP stack located at a front-end node and a selected back-end web server. A handoff protocol implemented by the loadable modules works within the kernel level of the existing TCP/IP code. As such, no changes to the existing TCP/IP code is necessary. The loadable modules at the front-end are able to select a back-end web server depending on the content of the web request, coordinate handing off TCP states, and forward packets to the back-end web server. Loadable modules at the selected back-end modify response packets going out to reflect the proper TCP state of the front-end node.

摘要翻译： 一种在通信网络中切断TCP状态的方法和系统。 具体地说，本发明允许在针对TCP状态的罕见切换优化的关联网络中的节点之间切换TCP状态。 切换发生在围绕位于前端节点的TCP / IP堆栈的动态可加载模块和选定的后端Web服务器之间。 由可加载模块实现的切换协议在现有TCP / IP代码的内核级别内工作。 因此，不需要对现有TCP / IP代码进行任何更改。 前端的可加载模块能够根据Web请求的内容选择后端Web服务器，协调处理TCP状态，并将数据包转发到后端Web服务器。 所选后端的可加载模块会修改响应数据包，以反映前端节点的正确TCP状态。

公开(公告)号：US07689710B2

公开(公告)日：2010-03-30

申请号：US09880632

申请日：2001-06-12

申请人： Wenting Tang ,  Ludmila Cherkasova ,  Lance Warren Russell

发明人： Wenting Tang ,  Ludmila Cherkasova ,  Lance Warren Russell

IPC分类号： G06F15/16 ,  G06F15/177 ,  G06F15/173

CPC分类号： H04L29/06 ,  H04L67/02 ,  H04L67/1002 ,  H04L67/1014 ,  H04L67/1027 ,  H04L67/142 ,  H04L69/16 ,  H04L69/163

摘要： A method and system for handing-off TCP states in a communication network. Specifically, the present invention discloses allows for transferring TCP states between front-end node and a plurality of back-end web servers. The handoff occurs between dynamically loadable modules that wrap around the TCP/IP stack located at a front-end node and a selected back-end web server. A handoff protocol implemented by the loadable modules works within the kernel level of the existing TCP/IP code. As such, no changes to the existing TCP/IP code is necessary. The loadable modules at the front-end are able to select a back-end web server depending on the HTTP request, coordinate handing off TCP states, and forward packets to the back-end web server. Loadable modules at the selected back-end modify response packets going out to reflect the proper TCP state of the front-end node.

摘要翻译： 一种在通信网络中切断TCP状态的方法和系统。 具体地，本发明公开了允许在前端节点和多个后端web服务器之间传送TCP状态。 切换发生在围绕位于前端节点的TCP / IP堆栈的动态可加载模块和选定的后端Web服务器之间。 由可加载模块实现的切换协议在现有TCP / IP代码的内核级别内工作。 因此，不需要对现有TCP / IP代码进行任何更改。 前端的可加载模块能够根据HTTP请求选择后端Web服务器，协调处理TCP状态，并将数据包转发到后端Web服务器。 所选后端的可加载模块会修改响应数据包，以反映前端节点的正确TCP状态。

公开(公告)号：US07924829B2

公开(公告)日：2011-04-12

申请号：US11742073

申请日：2007-04-30

申请人： Kevin Lynn Fought ,  Vinit Jain ,  Lance Warren Russell

发明人： Kevin Lynn Fought ,  Vinit Jain ,  Lance Warren Russell

IPC分类号： H04L12/28

CPC分类号： H04L45/745 ,  H04L41/5009 ,  H04L43/00 ,  H04L45/00

摘要： A computer implemented method, data processing system, and computer program product for enabling network statistics to be collected per software partition. When an incoming data packet is received at the Internet Protocol (IP) layer of an operating system, the IP address in the incoming data packet is identified. The IP address in the incoming data packet is then compared against IP addresses stored in a hash table, wherein the hash table maps each IP address in the hash table to a software partition of the operating system. If an IP address in the hash table matches the IP address in the data packet, the software partition mapped to the located IP address in the hash table is identified. Network statistics collected for the incoming data packet are then associated with the software partition.

摘要翻译： 计算机实现的方法，数据处理系统和计算机程序产品，用于每个软件分区收集网络统计信息。 当在操作系统的因特网协议（IP）层接收到输入数据分组时，识别输入数据分组中的IP地址。 然后将输入数据包中的IP地址与存储在散列表中的IP地址进行比较，其中散列表将散列表中的每个IP地址映射到操作系统的软件分区。 如果哈希表中的IP地址与数据包中的IP地址匹配，则会识别映射到哈希表中找到的IP地址的软件分区。 然后将收到的数据包的网络统计信息与软件分区相关联。

公开(公告)号：US20100325471A1

公开(公告)日：2010-12-23

申请号：US12486332

申请日：2009-06-17

申请人： Rajeev Mishra ,  Lance Warren Russell ,  Ravi A. Shankar

发明人： Rajeev Mishra ,  Lance Warren Russell ,  Ravi A. Shankar

IPC分类号： G06F11/14 ,  G06F12/16 ,  G06F9/455

CPC分类号： G06F11/1438 ,  G06F11/1484 ,  G06F11/203 ,  G06F11/2035

摘要： A computer implemented method, a tangible computer storage medium, and a data processing system provide high availability support for virtual machines in a logical partitioned platform. A monitoring system detect a failure in the virtual machine. Partition management firmware then restarts the virtual machine in a consistency failover image node utilizing a consistency failover image. If a subsequent failure of the virtual machine is detected within a predetermined time, partition management firmware restarts the virtual machine in a boot failover image node utilizing a boot failover image.

摘要翻译： 计算机实现的方法，有形计算机存储介质和数据处理系统为逻辑分区平台中的虚拟机提供高可用性支持。 监控系统检测虚拟机故障。 分区管理固件然后使用一致性故障转移映像重新启动一个一致故障转移映像节点中的虚拟机。 如果在预定时间内检测到虚拟机的后续故障，则分区管理固件使用引导故障切换映像重新启动引导故障转移映像节点中的虚拟机。

公开(公告)号：US07797576B2

公开(公告)日：2010-09-14

申请号：US11741500

申请日：2007-04-27

申请人： David Jones Craft ,  Vinit Jain ,  Lance Warren Russell

发明人： David Jones Craft ,  Vinit Jain ,  Lance Warren Russell

IPC分类号： G06F11/00

CPC分类号： G06F11/1438

摘要： A computer implemented method, apparatus, and computer program product for managing state data in a workload partitioned environment. Process state data for a process in a workload partition is saved. Process state data is used to restore the process to a given state. State data associated with open sockets and open files bound to the first process is saved. In response to a determination that the process is associated with a domain socket that is bound to a socket file, an absolute pathname for the socket file is saved. A domain socket associated with a connecting process in the workload partition uses the socket file to connect to a domain socket associated with a listening process in the same workload partition to enable the two processes to communicate with each other.

摘要翻译： 一种用于在工作负载分区环境中管理状态数据的计算机实现的方法，装置和计算机程序产品。 保存工作负载分区中进程的进程状态数据。 过程状态数据用于将进程恢复到给定状态。 保存与第一个进程绑定的打开的套接字和打开文件关联的状态数据。 响应确定该进程与绑定到套接字文件的域套接字相关联，将保存套接字文件的绝对路径名。 与工作负载分区中的连接进程相关联的域套接字使用套接字文件连接到与同一工作负载分区中的侦听进程相关联的域套接字，以使两个进程能够相互通信。

公开(公告)号：US20090248887A1

公开(公告)日：2009-10-01

申请号：US12057708

申请日：2008-03-28

申请人： David Jones Craft ,  Lance Warren Russell

发明人： David Jones Craft ,  Lance Warren Russell

IPC分类号： G06F15/16

CPC分类号： H04L67/06

摘要： The present invention provides a computer storage server implemented method for generating attribute streams. The storage server receives a request from an application to write data to at least one storage object, the request having a request type. The storage server then updates an attribute stream with metadata associated with the at least one storage object. The storage server also updates the attribute stream with the at least one checksum, such that the attribute stream is associated with the at least one storage object. The storage server determines the request type to be one selected from among the group consisting of an email type, a file transfer protocol type, and a hypertext transfer protocol type. Responsive to a determination that the request type is among one selected from the group consisting of an email type, a file transfer protocol type, and a hypertext transfer protocol type, the storage server verifies the application. Based on the determination of the request type, the storage server then updates the attribute stream by applying a filter among one selected from the group consisting of an email attribute stream filter, file transfer protocol attribute stream filter and a hypertext transfer protocol attribute stream filter.

摘要翻译： 本发明提供一种用于生成属性流的计算机存储服务器实现方法。 存储服务器从应用程序接收到向至少一个存储对象写入数据的请求，该请求具有请求类型。 然后，存储服务器使用与至少一个存储对象相关联的元数据来更新属性流。 存储服务器还使用至少一个校验和更新属性流，使得属性流与至少一个存储对象相关联。 存储服务器将请求类型确定为从由电子邮件类型，文件传输协议类型和超文本传输​​协议类型组成的组中选择的请求类型。 响应于确定请求类型是从由电子邮件类型，文件传输协议类型和超文本传输​​协议类型组成的组中选择的一种，存储服务器验证应用程序。 基于请求类型的确定，存储服务器然后通过在从由电子邮件属性流过滤器，文件传输协议属性流过滤器和超文本传输​​协议属性流过滤器组成的组中选择的一个中应用过滤器来更新属性流。

公开(公告)号：US20090094582A1

公开(公告)日：2009-04-09

申请号：US11867303

申请日：2007-10-04

申请人： David Jones Craft ,  Vinit Jain ,  Lance Warren Russell

发明人： David Jones Craft ,  Vinit Jain ,  Lance Warren Russell

IPC分类号： G06F9/44

CPC分类号： G06F9/4862

摘要： A computer implemented method, apparatus, and computer program product for checkpoint and restart of a process. In one embodiment, a process shell representing a restartable process is created on an arrival server in response to identifying a process on a departure server that is a restartable process. The process shell is frozen. A name of the restartable process, an identifier of the restartable process, and an identifier of a parent process of the restartable process is assigned to the process shell. The process shell is then unfrozen and exits. A parent process restarts the restartable process when the process shell exits.

摘要翻译： 一种计算机实现的方法，装置和计算机程序产品，用于检查点和重新启动过程。 在一个实施例中，响应于识别作为可重启过程的出发服务器上的过程，在到达服务器上创建表示可重启过程的进程外壳。 进程shell被冻结。 可重新启动的进程的名称，可重新启动的进程的标识符以及可重新启动进程的父进程的标识符被分配给进程外壳。 然后，进程shell解冻并退出。 当进程shell退出时，父进程重新启动可重新启动的进程。

公开(公告)号：US20080295111A1

公开(公告)日：2008-11-27

申请号：US11751505

申请日：2007-05-21

申请人： DAVID JONES CRAFT ,  Rajeev Mishra ,  Lance Warren Russell

发明人： DAVID JONES CRAFT ,  Rajeev Mishra ,  Lance Warren Russell

IPC分类号： G06F3/00

CPC分类号： G06F11/1438 ,  G06F11/1407

摘要： A computer implemented method, apparatus, and computer program product for a checkpoint process associated with a device driver in a workload partitioned environment. In response to initiation of a checkpoint process, a stream is frozen. The stream comprises a set of kernel modules driving a device. Freezing the stream prevents any module in the set of kernel modules from sending any messages, other than a checkpoint message, to another module in the set of kernel modules. The message block for each module in the set of kernel modules is updated with internal data to form a restart message. The internal data is data describing a state of the module in the set of kernel modules.

摘要翻译： 一种用于与工作负载分区环境中的设备驱动程序相关联的检查点处理的计算机实现的方法，装置和计算机程序产品。 响应启动检查点过程，流被冻结。 该流包括驱动设备的一组内核模块。 冻结流可防止内核模块集中的任何模块发送除检查点消息之外的任何消息到内核模块集中的另一个模块。 内核模块集中的每个模块的消息块将使用内部数据进行更新，以形成重新启动消息。 内部数据是描述内核模块集中模块状态的数据。

公开(公告)号：US07346748B2

公开(公告)日：2008-03-18

申请号：US11117143

申请日：2005-04-28

申请人： Matthew Albert Huras ,  Thomas Stanley Mathews ,  Lance Warren Russell

发明人： Matthew Albert Huras ,  Thomas Stanley Mathews ,  Lance Warren Russell

IPC分类号： G06F13/00

CPC分类号： G06F17/30067

摘要： A method, apparatus, and computer program product for storage pools with write atomicity. An abstraction manager enforces write atomicity and disallows options which are inconsistent with write atomicity. The abstraction manager constructs through a physical device interface a logical continuous view of a storage pool in a manner consistent with write atomicity. Applications collect information specific to write atomicity from the abstraction manager through an application interface.

公开(公告)号：US5892753A

公开(公告)日：1999-04-06

申请号：US755867

申请日：1996-12-02

申请人： Daniel Edward Badt ,  David Richard Marquardt ,  Satyanarayana Billekal Rao ,  Lance Warren Russell ,  Larry Steven Wise

发明人： Daniel Edward Badt ,  David Richard Marquardt ,  Satyanarayana Billekal Rao ,  Lance Warren Russell ,  Larry Steven Wise

IPC分类号： H04L12/56

CPC分类号： H04L47/10 ,  H04L47/36

摘要： A local system issues an ICMP echo request packet with a size based on an initial PMTU value corresponding to that of the local system's interface MTU. When an ICMP needfrag error packet is received by the local system, this signifies that the initial PMTU value was too large for packets of corresponding size to be transmitted through the Internet without fragmentation. The value is accordingly decreased by the local system and a next ICMP echo request packet transmitted through the Internet. The process iterates until an ICMP echo request packet is successfully transmitted through the Internet to a target system. In response thereto, the target system transmits an ICMP echo response packet back through the Internet to the local system, thereby indicating that the last-sent ICMP echo request packet from the local system had the correct PMTU value which is thereafter assumed to be the correct PMTU. An MTU application, MTU table, and corresponding application program interfaces and controls are provided in the system for dynamically maintaining PMTU value estimates and for providing next estimates of PMTU. These next estimates are accessed from the table in next ICMP echo request packet transmissions when an ICMP needfrag error packet does not contain an MTU value returned from a router in the Internet.

摘要翻译： 本地系统发布基于与本地系统接口MTU对应的初始PMTU值的ICMP回显请求报文。 当本地系统接收到ICMP需要的报文错误报文时，表示初始的PMTU值对于相应大小的报文来说太大，无法通过因特网进行传输。 因此本地系统和通过因特网传输的下一个ICMP回显请求报文相应地减小了该值。 该过程迭代直到ICMP回显请求分组通过因特网成功传输到目标系统。 响应于此，目标系统通过因特网向本地系统发送ICMP回应响应分组，从而指示来自本地系统的最后发送的ICMP回应请求分组具有正确的PMTU值，其后被认为是正确的 PMTU。 在系统中提供MTU应用，MTU表和相应的应用程序接口和控制，用于动态维护PMTU值估计和提供PMTU的下一个估计。 当ICMP需要帧错误分组不包含从因特网中的路由器返回的MTU值时，从下一个ICMP回显请求分组传输中的表中访问这些下一个估计。