公开(公告)号：US20150373115A1

公开(公告)日：2015-12-24

申请号：US14747421

申请日：2015-06-23

Applicant: Liqid Inc.

Inventor： Jason Breakstone ,  Christopher R. Long ,  James Scott Cannata

IPC: H04L29/08 ,  G06F13/42 ,  G06F13/40

CPC classification number: G06F11/2053 ,  G06F11/2007 ,  G06F11/2012 ,  G06F11/2017 ,  G06F11/2069 ,  G06F13/4022 ,  G06F13/4068 ,  G06F13/4234 ,  G06F13/4282 ,  G06F2201/85 ,  H04L67/1097 ,  H04L69/40

Abstract: Systems, methods, apparatuses, and software for data storage systems are provided herein. In one example, a data storage system is provided that includes a plurality of storage drives each comprising a Peripheral Component Interconnect Express (PCIe) interface, and configured to store data and retrieve the data stored on associated storage media responsive to storage operations. The data storage system includes one or more processing modules comprising one or more processors communicatively coupled to the plurality of storage drives over a PCIe fabric comprised of one or more PCIe switches. The processors are configured to share a PCIe address space associated with the PCIe fabric for transfer of the storage operations to appropriate ones of the processors that manage ones of the plurality of storage drives.

Abstract translation: 本文提供了用于数据存储系统的系统，方法，装置和软件。 在一个示例中，提供了数据存储系统，其包括多个存储驱动器，每个存储驱动器包括外围组件互连Express（PCIe）接口，并且被配置为响应于存储操作来存储数据并检索存储在相关联的存储介质上的数据。 数据存储系统包括一个或多个处理模块，其包括通过由一个或多个PCIe交换机组成的PCIe结构通信地耦合到多个存储驱动器的一个或多个处理器。 处理器被配置为共享与PCIe结构相关联的PCIe地址空间，用于将存储操作传送到管理多个存储驱动器中的一个的适当处理器。

公开(公告)号：US20150358196A1

公开(公告)日：2015-12-10

申请号：US14806750

申请日：2015-07-23

Applicant: SAS Institute Inc.

Inventor： Gerald Donald Baulier ,  Vincent L. Deters ,  Scott J. Kolodzieski

IPC: H04L12/24 ,  H04L12/58

CPC classification number: H04L41/0663 ,  G06F9/542 ,  G06F11/2007 ,  G06F11/2028 ,  G06F11/2041 ,  G06F2201/85 ,  H04L51/18

Abstract: In a system, a first status of a first ESP engine (ESPE) executing at a first computing device is determined as newly active; a last published event block object identifier is determined as an identifier uniquely identifying a last event block object published to an out-messaging network device; a next event block object having an event block object identifier greater than the determined last published event block object identifier is selected from a first computer-readable medium; and the selected next event block object is published to the out-messaging network device. A first event block object is received from a second ESPE executing at a second computing device. A first status of the second ESPE is determined as standby by the second computing device. The received first event block object is stored by the second computing device in a second non-transitory computer-readable medium.

Abstract translation: 在系统中，在第一计算设备处执行的第一ESP引擎（ESPE）的第一状态被确定为新活动的; 确定最后发布的事件块对象标识符作为唯一地标识发布到外发消息传递网络设备的最后事件块对象的标识符; 从第一计算机可读介质中选择具有大于所确定的最后发布的事件块对象标识符的事件块对象标识符的下一个事件块对象; 并且所选择的下一个事件块对象被发布到外发消息网络设备。 从在第二计算设备处执行的第二ESPE接收第一事件块对象。 第二ESPE的第一状态被第二计算设备确定为待机。 所接收的第一事件块对象由第二计算设备存储在第二非暂时计算机可读介质中。

公开(公告)号：US20150350374A1

公开(公告)日：2015-12-03

申请号：US14820657

申请日：2015-08-07

Applicant: FUJITSU LIMITED

Inventor： Tatsuya Soneda ,  Tetsu Yamamoto ,  NORIHITO NISHIMOTO ,  Tadashige IWAO

IPC: H04L29/08 ,  G06F11/20

CPC classification number: H04L67/32 ,  G06F11/2033 ,  G06F2201/85 ,  H04L67/10 ,  H04W40/02 ,  H04W88/04

Abstract: A selection device is selected from a first group in a first cluster. The node devices of the first group can communicate with a node device in a second cluster adjacent to the first cluster. The selection device performs transmission and reception of a report frame that reports an identifier of a node device included in the first group. The selection device selects from the first group a first relay device that relays a relay frame used for a communication between a node device in the first cluster and a node device in the second cluster. The first relay device determines anode device adjacent to the first relay device to be a second relay device that relays the relay frame to a node device in the second cluster.

Abstract translation: 从第一集群中的第一组中选择选择设备。 第一组的节点设备可以与第一集群相邻的第二集群中的节点设备进行通信。 选择装置执行报告帧的发送和接收，该报告帧报告包括在第一组中的节点装置的标识符。 选择装置从第一组中选择中继用于第一集群中的节点装置与第二集群中的节点装置之间的通信的中继帧的第一中继装置。 第一中继装置确定与第一中继装置相邻的阳极装置是将中继帧中继到第二集群中的节点装置的第二中继装置。

公开(公告)号：US09152513B2

公开(公告)日：2015-10-06

申请号：US14160264

申请日：2014-01-21

Applicant: NetApp, Inc.

Inventor： Mayank Saxena ,  Melvin McGee

IPC: G06F11/00 ,  G06F11/20 ,  H04L12/707 ,  G06F13/42

CPC classification number: G06F11/201 ,  G06F11/0727 ,  G06F11/0793 ,  G06F11/2012 ,  G06F11/2089 ,  G06F13/426 ,  G06F2201/805 ,  G06F2201/85 ,  H04L45/22 ,  H04L47/35

Abstract: Technology is disclosed for recovering I/O modules in a storage system using in-band alternate control path (ACP) architecture (“the technology”). The technology enables a storage server to transmit control commands, e.g., for recovering an I/O module, to the I/O module over a data path that is typically used to transmit data commands. The control commands are typically transmitted using ACP that is separate from the data path. By enabling transmission of control commands over the data path, the technology eliminates the need for separate medium for ACP, at least in part, to transmit the control commands. The technology can be implemented in a pure in-band ACP mode, which supports recovering an I/O module of a storage shelf in which at least one I/O module is responsive, and/or in a mixed in-band ACP mode, which supports recovery of I/O modules of a storage shelf in which all I/O modules are non-responsive.

Abstract translation: 公开了用于使用带内替代控制路径（ACP）架构（“该技术”）在存储系统中恢复I / O模块的技术。 该技术使得存储服务器能够通过通常用于发送数据命令的数据路径向I / O模块发送例如用于恢复I / O模块的控制命令。 控制命令通常使用与数据路径分离的ACP传输。 通过使数据路径上的控制命令能够传输，该技术消除了对于ACP的单独介质的需要，至少部分地传送控制命令。 该技术可以以纯带内ACP模式实现，其支持恢复其中至少一个I / O模块响应的存储架的I / O模块和/或混合带内ACP模式， 其支持对所有I / O模块不响应的存储架的I / O模块的恢复。

公开(公告)号：US20150227443A1

公开(公告)日：2015-08-13

申请号：US14405377

申请日：2013-10-31

Applicant: Hangzhou H3C Technologies Co., Ltd.

Inventor： Qiang Wu

IPC: G06F11/20 ,  G06F13/40

CPC classification number: G06F11/2007 ,  G06F13/4022 ,  G06F2201/85 ,  H04L12/4625 ,  H04L12/4641

Abstract: According to an example, an Edge Virtual Bridging (EVB) station is configured with a VM, an ER and multiple physical network cards. The VM is configured with multiple virtual network cards and each virtual network card has one VSI. Each VSI is connected with one of the physical network cards via the ER. One of the physical network cards is configured as a primary physical network card, and another is configured as a secondary physical network card. A VSI corresponding to the primary physical network card is configured as a primary virtual interface, and a VSI corresponding to the secondary physical network card is configured as a secondary virtual interface. After determining the primary physical network card failed, the secondary physical network card is configured as a new primary physical network card, and the secondary virtual is configured as a new primary virtual interface.

Abstract translation: 根据一个示例，边缘虚拟桥接（EVB）站配置有VM，ER和多个物理网卡。 VM配置有多个虚拟网卡，每个虚拟网卡都有一个VSI。 每个VSI通过ER连接到一个物理网卡。 其中一个物理网卡被配置为主物理网卡，另一个配置为辅助物理网卡。 将与主要物理网卡相对应的VSI配置为主虚拟接口，将与辅助物理网卡对应的VSI配置为辅助虚拟接口。 确定主要物理网卡失败后，将备用物理网卡配置为新的主要物理网卡，将辅助虚拟机配置为新的主虚拟接口。

公开(公告)号：US20150212907A1

公开(公告)日：2015-07-30

申请号：US14163558

申请日：2014-01-24

Applicant: International Business Machines Corporation

Inventor： Bertrand H. Portier ,  Suraj Shinde

IPC: G06F11/20 ,  G06F17/30

CPC classification number: G06F11/2007 ,  G06F11/0709 ,  G06F11/079 ,  G06F11/203 ,  G06F11/2035 ,  G06F11/2048 ,  G06F17/30575 ,  G06F2201/85

Abstract: Embodiments of the present invention disclose a method, computer program product, and system for memory replication. In one embodiment, in accordance with the present invention, the computer implemented method includes the steps of executing a mobile agent on a server node, wherein the server node is within a cluster of server nodes connected via network communications, capturing a memory state of the server node during operation of the server node, wherein the memory state includes session information stored on computer memory of the server node, which is captured and stored by the mobile agent, monitoring the server node to determine whether the server node has failed, and responsive to determining that the server node has failed, migrating the mobile agent to an active server node within the cluster of server nodes, wherein the mobile agent carries the captured memory state.

Abstract translation: 本发明的实施例公开了一种用于存储器复制的方法，计算机程序产品和系统。 在一个实施例中，根据本发明，计算机实现的方法包括以下步骤：在服务器节点上执行移动代理，其中服务器节点在通过网络通信连接的服务器节点的集群内，捕获 服务器节点，其中所述存储器状态包括存储在所述服务器节点的计算机存储器上的会话信息，所述会话信息被所述移动代理捕获和存储，监视所述服务器节点以确定所述服务器节点是否已经失败，并且响应 以确定服务器节点已经失败，将移动代理迁移到服务器节点的集群内的活动服务器节点，其中移动代理携带捕获的存储器状态。

公开(公告)号：US20150149823A1

公开(公告)日：2015-05-28

申请号：US14316273

申请日：2014-06-26

Applicant: NetApp, Inc.

Inventor： Brian McCarthy ,  Donna Duffy ,  Joseph CaraDonna

IPC: G06F11/14

CPC classification number: G06F11/1471 ,  G06F11/0712 ,  G06F11/1438 ,  G06F11/1441 ,  G06F2201/805 ,  G06F2201/815 ,  G06F2201/85

Abstract: A method, non-transitory computer readable medium, and host device that receives one or more transactions. A state is stored in a transaction log in a volatile memory wherein the state includes information associated with the one or more transactions. The transaction log is stored in a stable storage device when a failure is determined to a have occurred. The transaction log can then be retrieved and replayed subsequent to a reboot. Thereby, state can be preserved and transactions pending, but not yet committed to storage server devices, can be replayed and proceed with minimal or no impact on the client devices originating the write transactions.

Abstract translation: 接收一个或多个事务的方法，非暂时计算机可读介质和主机设备。 状态存储在易失性存储器中的事务日志中，其中状态包括与一个或多个事务相关联的信息。 当确定发生故障时，事务日志存储在稳定的存储设备中。 然后可以在重新启动后检索和重播事务日志。 因此，可以保留状态，并且可以重放事件待处理但尚未提交到存储服务器设备，并且对发起写事务的客户端设备进行最小或不影响。

公开(公告)号：US20150149814A1

公开(公告)日：2015-05-28

申请号：US14555285

申请日：2014-11-26

Applicant: Futurewei Technologies, Inc.

Inventor： Da Qi Ren ,  Zhulin Wei

IPC: G06F11/20 ,  G06F11/14 ,  H04L29/08 ,  H04L12/26

CPC classification number: G06F11/2028 ,  G06F11/0703 ,  G06F11/1438 ,  G06F11/1448 ,  G06F11/1484 ,  G06F11/203 ,  G06F2201/84 ,  G06F2201/85 ,  H04L43/10 ,  H04L67/1002 ,  H04L67/1095

Abstract: A method of providing failure recovery capabilities to a cloud environment for scientific HPC applications. An HPC application with MPI implementation extends the class of MPI programs to embed the HPC application with various degrees of fault tolerance. An MPI fault tolerance mechanism realizes a recover-and-continue solution. If an error occurs, only failed processes re-spawn, the remaining living processes remain in their original processors/nodes, and system recovery costs are thus minimized.

Abstract translation: 为科学HPC应用程序向云环境提供故障恢复功能的方法。 具有MPI实现的HPC应用程序扩展了MPI程序类，以嵌入具有不同程度容错能力的HPC应用程序。 MPI容错机制实现了恢复和持续解决方案。 如果发生错误，只有失败的进程重新生成，剩余的生存过程将保留在其原始处理器/节点中，因此系统恢复成本最小化。

公开(公告)号：US08984334B2

公开(公告)日：2015-03-17

申请号：US13631272

申请日：2012-09-28

Applicant: Kazuhisa Fukuda

Inventor： Kazuhisa Fukuda

IPC: G06F11/00 ,  G06F11/07 ,  G06F9/48 ,  G06F11/20

CPC classification number: G06F11/2028 ,  G06F9/4881 ,  G06F11/0715 ,  G06F11/0793 ,  G06F11/203 ,  G06F11/2035 ,  G06F11/2043 ,  G06F2201/805 ,  G06F2201/82 ,  G06F2201/85

Abstract: A processor includes a plurality of processing sections, each of which executes a predetermined process. A plurality of fault detecting circuits are respectively provided for the plurality of processing sections, to detect a fault in one of the plurality of processing sections as a fault processing section to generate a fault detection signal. A fault monitoring and control section controls a normal processing section as at least one of the plurality of processing sections other than the fault processing section to execute a relieving process in response to the fault detection signal. The relieving process is determined based on a process load of the fault processing section, a process load of the normal processing section, and priority levels of processes to be executed by the fault processing section and the normal processing section.

Abstract translation: 处理器包括多个处理部分，每个处理部分执行预定处理。 分别为多个处理部提供多个故障检测电路，以将多个处理部中的一个处理部的故障检测为故障处理部，生成故障检测信号。 故障监视和控制部分控制除了故障处理部分之外的多个处理部分中的至少一个的正常处理部分，以响应故障检测信号执行缓解处理。 基于故障处理部的处理负荷，正常处理部的处理负荷以及由故障处理部和正常处理部执行的处理的优先级来确定解除处理。

公开(公告)号：US20140359344A1

公开(公告)日：2014-12-04

申请号：US14459890

申请日：2014-08-14

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Hao HE ,  Ling SUN ,  Jun Wei ZHANG

IPC: G06F11/20 ,  H04L29/08

CPC classification number: G06F11/2007 ,  G06F11/2017 ,  G06F11/2028 ,  G06F11/2038 ,  G06F11/2046 ,  G06F11/2089 ,  G06F11/2094 ,  G06F2201/85 ,  H04L41/0668 ,  H04L41/0816 ,  H04L67/1002 ,  H04L67/1097 ,  H04L69/40

Abstract: An SVC cluster manages a plurality of storage devices and includes a plurality of SVCs interconnected via a network, each SVC acting as a separate node. A new configuration node is activated in response to configuration node failures. The new configuration node retrieves client subscription information about events occurring in storage devices managed by the SVC cluster from the storage devices. In response to events occurring in the storage device managed by the SVC cluster, the new configuration node obtains storage device event information from a storage device event monitoring unit. The new configuration node sends storage device events to clients who have subscribed to this information according to subscription information obtained. The storage device is not installed in the original configuration node.

Abstract translation: SVC集群管理多个存储设备并且包括经由网络互连的多个SVC，每个SVC充当单独的节点。 响应于配置节点故障激活新的配置节点。 新配置节点从存储设备检索关于由SVC集群管理的存储设备中发生的事件的客户端订阅信息。 响应于由SVC集群管理的存储设备中发生的事件，新的配置节点从存储设备事件监控单元获取存储设备事件信息。 新配置节点根据获得的订阅信息向已经订阅该信息的客户端发送存储设备事件。 存储设备未安装在原始配置节点中。