公开(公告)号：US20080049617A1

公开(公告)日：2008-02-28

申请号：US11466615

申请日：2006-08-23

申请人： Donald G. Grice ,  Kalyan C. Gunda ,  Brian D. Herr ,  Gautam H. Shah

发明人： Donald G. Grice ,  Kalyan C. Gunda ,  Brian D. Herr ,  Gautam H. Shah

IPC分类号： H04J1/16

CPC分类号： H04L47/10 ,  H04L47/12 ,  H04L47/17 ,  H04L47/32

摘要： A system for flow-control concurrency to prevent excessive packet loss, including at least one transmitter node. Each transmitter node is configured to transmit data. A first flow-control device is coupled to the at least one transmitter node. The first flow-control device is configured to limit the number of concurrent data replies sent by the at least one transmitter node such that the resources on the transmitter node side will not be overrun. At least one receive node is configured to receive data transmitted. The at least one receiver node is coupled to the at least one transmitter node via the communication network. A second flow-control device is coupled to the at least one receiver node. The second flow-control device is configured to limit the number of concurrent data requests received by the at least one receiver node such that the resources on the receiver node side will not be overrun.

摘要翻译： 一种用于流控并发的系统，用于防止过多的分组丢失，包括至少一个发射机节点。 每个发射机节点被配置为发送数据。 第一流量控制装置耦合到所述至少一个发射器节点。 第一流量控制装置被配置为限制由至少一个发射机节点发送的并发数据回复的数量，使得发射机节点侧的资源不会超载。 至少一个接收节点被配置为接收发送的数据。 所述至少一个接收器节点经由所述通信网络耦合到所述至少一个发射机节点。 第二流量控制装置耦合到所述至少一个接收器节点。 第二流量控制装置被配置为限制由至少一个接收器节点接收的并发数据请求的数量，使得接收机节点侧的资源不会超载。

公开(公告)号：US07302607B2

公开(公告)日：2007-11-27

申请号：US10652509

申请日：2003-08-29

申请人： Kalyan C. Gunda ,  Brian D. Herr

发明人： Kalyan C. Gunda ,  Brian D. Herr

IPC分类号： G06F11/00

CPC分类号： G06F11/2092 ,  G06F11/2033

摘要： A method for recovery in a two-node data processing system is provided wherein each node is a primary server for a first nonvolatile storage device and for which there is provided shared access to a second nonvolatile storage device for which the other node is a primary server and wherein each node also includes a direct connection to the shared nonvolatile storage device for which the other node is the primary server. Upon notification of failure, the method operates by first confirming continued access by each node to the nonvolatile storage device for which it is the primary server and then by attempting to access the shared nonvolatile storage device via the direct connection and by waiting for a time sufficient for the same process to be carried out by the other node. If access to the shared nonvolatile storage device is successful, the node takes control of both nonvolatile storage devices. If the access is not successful a comparison of node numbers is carried out to decide the issue of control. Whenever a node determines that it does not have access to the storage device for which it is the primary server, it shuts down recovery at the node.

摘要翻译： 提供了一种用于在双节点数据处理系统中进行恢复的方法，其中每个节点是用于第一非易失性存储设备的主服务器，并且提供对第二非易失性存储设备的共享访问，其中另一个节点是主服务器 并且其中每个节点还包括到所述共享非易失性存储设备的直接连接，所述另一个节点是主服务器。 在通知故障时，该方法通过首先确认每个节点对其作为主服务器的非易失性存储设备的持续访问，然后通过直接连接尝试访问共享的非易失性存储设备，并且等待足够的时间 对于由另一个节点执行的相同进程。 如果访问共享的非易失性存储设备成功，则节点控制两个非易失性存储设备。 如果访问不成功，则执行节点号的比较来决定控制的问题。 每当一个节点确定它无法访问作为主服务器的存储设备时，它将关闭该节点的恢复。

公开(公告)号：US07093097B2

公开(公告)日：2006-08-15

申请号：US09994443

申请日：2001-11-27

申请人： Brian D. Herr ,  Joseph J. Kavaky

发明人： Brian D. Herr ,  Joseph J. Kavaky

IPC分类号： G06F12/00

CPC分类号： G06F5/065 ,  G06F2205/064

摘要： A method and system for maintaining a pinned kernel memory pool from which memory buffers are allocated to time critical processing tasks. The memory pool initially comprises a base block of pinned kernel memory to which additional memory blocks are added as more buffers are requested. Non-contiguous memory blocks may be used because they are maintained as a linked list. New buffer requests are satisfied from the earliest added blocks. The most recently added memory blocks are periodically examined for use to determine if they are in use and are returned to the kernel memory pool if they are unused.

公开(公告)号：US07890555B2

公开(公告)日：2011-02-15

申请号：US11775381

申请日：2007-07-10

申请人： Kalyan C. Gunda ,  Brian D. Herr ,  Frank B. Schmuck

发明人： Kalyan C. Gunda ,  Brian D. Herr ,  Frank B. Schmuck

IPC分类号： G06F12/00 ,  G06F17/30

CPC分类号： G06F11/004 ,  G06F11/1425 ,  G06F11/1471 ,  G06F11/202 ,  G06F11/2046

摘要： A method effectively preventing a requesting node from unfencing and mounting a file system subsequent to a failure in a cluster file system having a plurality of active nodes. The method comprising first upgrading one active node in the cluster to function as a cluster manager node. The cluster manager is in communication with all nodes. The cluster manager is assigned manager responsibilities, in part, comprising first receiving an active status request from the node requesting to mount a file system. The cluster manager first queries the quorum nodes to determine whether each node considers the cluster manager to still have cluster management responsibilities for the file system. If a majority of quorum nodes consider the cluster manager to still have cluster management responsibilities for the file system then the cluster manager responds to the requesting node's active status request. Thereafter, the requesting node proceeds with mounting the file system.

摘要翻译： 一种有效地防止请求节点在具有多个活动节点的集群文件系统中的故障之后对文件系统进行打包和装载的方法。 该方法包括首先升级集群中的一个主动节点以用作集群管理器节点。 集群管理器与所有节点通信。 分配管理器的职责是部分地包括首先从请求装载文件系统的节点接收活动状态请求。 群集管理器首先查询仲裁节点，以确定每个节点是否认为群集管理器仍然具有文件系统的群集管理职责。 如果大多数仲裁节点认为集群管理器仍然对文件系统具有集群管理职责，则集群管理器将响应请求节点的活动状态请求。 此后，请求节点继续安装文件系统。

公开(公告)号：US07533295B2

公开(公告)日：2009-05-12

申请号：US11772293

申请日：2007-07-02

申请人： Kalyan C. Gunda ,  Brian D. Herr

发明人： Kalyan C. Gunda ,  Brian D. Herr

IPC分类号： G06F11/00

CPC分类号： G06F11/2092 ,  G06F11/2033

摘要： A method for recovery in a two-node data processing system is provided wherein each node is a primary server for a first nonvolatile storage device and for which there is provided shared access to a second nonvolatile storage device for which the other node is a primary server and wherein each node also includes a direct connection to the shared nonvolatile storage device for which the other node is the primary server. Upon notification of failure, the method operates by first confirming continued access by each node to the nonvolatile storage device for which it is the primary server and then by attempting to access the shared nonvolatile storage device via the direct connection and by waiting for a time sufficient for the same process to be carried out by the other node. If access to the shared nonvolatile storage device is successful, the node takes control of both nonvolatile storage devices. If the access is not successful a comparison of node numbers is carried out to decide the issue of control. Whenever a node determines that it does not have access to the storage device for which it is the primary server, it shuts down recovery at the node.

摘要翻译： 提供了一种用于在双节点数据处理系统中进行恢复的方法，其中每个节点是用于第一非易失性存储设备的主服务器，并且提供对第二非易失性存储设备的共享访问，其中另一个节点是主服务器 并且其中每个节点还包括到所述共享非易失性存储设备的直接连接，所述另一个节点是主服务器。 在通知故障时，该方法通过首先确认每个节点对其作为主服务器的非易失性存储设备的持续访问，然后通过直接连接尝试访问共享的非易失性存储设备，并且等待足够的时间 对于由另一个节点执行的相同进程。 如果访问共享的非易失性存储设备成功，则节点控制两个非易失性存储设备。 如果访问不成功，则执行节点号的比较来决定控制的问题。 每当一个节点确定它无法访问作为主服务器的存储设备时，它将关闭该节点的恢复。

公开(公告)号：US20090019098A1

公开(公告)日：2009-01-15

申请号：US11775381

申请日：2007-07-10

申请人： Kalyan C. Gunda ,  Brian D. Herr ,  Frank B. Schmuck

发明人： Kalyan C. Gunda ,  Brian D. Herr ,  Frank B. Schmuck

IPC分类号： G06F12/00

CPC分类号： G06F11/004 ,  G06F11/1425 ,  G06F11/1471 ,  G06F11/202 ,  G06F11/2046

摘要： A method effectively preventing a requesting node from unfencing and mounting a file system subsequent to a failure in a cluster file system having a plurality of active nodes. The method comprising first upgrading one active node in the cluster to function as a cluster manager node. The cluster manager is in communication with all nodes. The cluster manager is assigned manager responsibilities, in part, comprising first receiving an active status request from the node requesting to mount a file system. The cluster manager first queries the quorum nodes to determine whether each node considers the cluster manager to still have cluster management responsibilities for the file system. If a majority of quorum nodes consider the cluster manager to still have cluster management responsibilities for the file system then the cluster manager responds to the requesting node's active status request. Thereafter, the requesting node proceeds with mounting the file system.

摘要翻译： 一种有效地防止请求节点在具有多个活动节点的集群文件系统中的故障之后对文件系统进行打包和装载的方法。 该方法包括首先升级集群中的一个主动节点以用作集群管理器节点。 集群管理器与所有节点通信。 分配管理器的职责是部分地包括首先从请求装载文件系统的节点接收活动状态请求。 群集管理器首先查询仲裁节点，以确定每个节点是否认为群集管理器仍然具有文件系统的群集管理职责。 如果大多数仲裁节点认为集群管理器仍然对文件系统具有集群管理职责，则集群管理器将响应请求节点的活动状态请求。 此后，请求节点继续安装文件系统。