公开(公告)号：US06516391B1

公开(公告)日：2003-02-04

申请号：US09523737

申请日：2000-03-13

申请人： Yuji Tsushima ,  Hideya Akashi ,  Keitaro Uehara ,  Naoki Hamanaka ,  Toru Shonai ,  Tetsuhiko Okada ,  Masamori Kashiyama

发明人： Yuji Tsushima ,  Hideya Akashi ,  Keitaro Uehara ,  Naoki Hamanaka ,  Toru Shonai ,  Tetsuhiko Okada ,  Masamori Kashiyama

IPC分类号： G06F1200

CPC分类号： G06F12/0813 ,  G06F12/0817 ,  G06F15/177 ,  G06F2212/2542

摘要： In a multiprocessor arranged in accordance with either NUMA or UMA in which a plurality of processor nodes containing a plurality of processor units are coupled to each other via a network, a cache snoop operation executed in connection with a memory access operation is performed at two stages, namely, local snoop operation executed within a node, and global snoop operation among nodes. Before executing the local snoop operation, an ACTV command for specifying only an RAS of a memory is issued to a target node having a memory to be accessed, and the memory access is activated in advance. A CAS of a memory is additionally specified and a memory access is newly executed after the ACTV command has been issued and then a memory access command has been issued. When there is such a possibility that a memory to be accessed is cached in a processor node except for a source node, this memory access command is issued to be distributed to all nodes so as to execute the global snoop operation. On the other hand, when there is no possibility that the memory to be accessed is cached, this memory access command is transferred only to the target node in yan one-to-one correspondence.

摘要翻译： 在根据其中包含多个处理器单元的多个处理器节点经由网络彼此耦合的NUMA或UMA而布置的多处理器中，结合存储器访问操作执行的高速缓存侦听操作在两个阶段 即在节点内执行的本地侦听操作，以及节点之间的全局侦听操作。 在执行本地侦听操作之前，向具有存储器的目标节点发出用于指定存储器的RAS的ACTV命令，并且预先激活存储器访问。 另外指定存储器的CAS，并且在发出ACTV命令之后重新执行存储器访问，然后发出存储器访问命令。 当存在待访问的存储器存在除了源节点之外的处理器节点的可能性时，该存储器访问命令被发布以分发给所有节点，以便执行全局侦听操作。 另一方面，当不存在要访问的存储器被缓存时，该存储器访问命令仅以一对一对应的方式传送到目标节点。

公开(公告)号：US08954963B2

公开(公告)日：2015-02-10

申请号：US13368634

申请日：2012-02-08

申请人： Naoya Hattori ,  Yuta Sawa ,  Keitaro Uehara

发明人： Naoya Hattori ,  Yuta Sawa ,  Keitaro Uehara

IPC分类号： G06F9/455 ,  G06F9/44

CPC分类号： G06F9/4411 ,  G06F9/45558 ,  G06F2009/45575

摘要： A method of controlling a virtual machine of a computer, the computer comprising: a physical machine comprising an I/O adapter having a physical function that creates a virtual function; a virtualization unit that provides computer resources of the physical machine to the virtual machine; and an OS that is executed on the virtual machine, the virtualization unit creating a virtual machine to which the virtual function is assigned, the virtual machine running the OS thereon, the method comprising: a first step of detecting, by the virtualization unit, a state change of the I/O adapter; a second step of identifying, by the virtualization unit, when a state of the I/O adapter becomes a predetermined state, the virtual machine to which the virtual function is assigned; and a third step of notifying, by the virtualization unit, the OS running on the identified virtual machine of the state of the I/O adapter.

摘要翻译： 一种控制计算机的虚拟机的方法，所述计算机包括：物理机器，其包括具有产生虚拟功能的物理功能的I / O适配器; 一个向虚拟机提供物理机的计算机资源的虚拟化单元; 以及在虚拟机上执行的操作系统，所述虚拟化单元创建分配有所述虚拟功能的虚拟机，所述虚拟机在其上运行所述操作系统，所述方法包括：第一步骤，由所述虚拟化单元检测 状态改变的I / O适配器; 第二步骤，由所述虚拟化单元识别所述I / O适配器的状态何时变为预定状态，所述虚拟功能被分配给所述虚拟机; 以及第三步骤，通过虚拟化单元通知在所识别的虚拟机上运行的I / O适配器的状态的OS。

公开(公告)号：US20110238817A1

公开(公告)日：2011-09-29

申请号：US12980606

申请日：2010-12-29

申请人： Hideki OKITA ,  Masahiro Yoshizawa ,  Keitaro Uehara

发明人： Hideki OKITA ,  Masahiro Yoshizawa ,  Keitaro Uehara

IPC分类号： G06F15/173

CPC分类号： H04L43/0817 ,  H04L41/142

摘要： A failure factor can be promptly carved up by effectively prioritizing the ports to which MEPs are assigned in a condition where the maximum number of MEPs is limited. A network monitoring server monitors a network having plural switch devices. The plural switch devices each have a communication confirmation function using transmission and reception of a monitoring frame between ports. An upper limit is set to the number of ports to which maintenance end points (MEPs) are assignable among the respective ports of the plural switch devices. The network monitoring server calculates, for each of the plural switch devices, a monitoring priority corresponding to the probability of failures of a device connected to a network path including the port or the port, and generates data for displaying the calculated monitoring priority of each port in association with each port.

摘要翻译： 通过在限制MEP最大数量的条件下有效优先分配MEP分配的端口，可以及时刻录故障因素。 网络监视服务器监视具有多个交换设备的网络。 多个开关装置各自具有使用端口之间的监视帧的发送和接收的通信确认功能。 对多个开关装置的各个端口中的维护端点（MEP）可分配的端口数设定上限。 所述网络监视服务器针对所述多个交换机装置中的每一个计算与连接到包括所述端口或所述端口的网络路径的设备的故障概率相对应的监视优先级，并且生成用于显示所计算的每个端口的监视优先级的数据 与每个港口相关联。

公开(公告)号：US08352665B2

公开(公告)日：2013-01-08

申请号：US12486927

申请日：2009-06-18

申请人： Keishi Nakayama ,  Keitaro Uehara ,  Takashi Aoyagi ,  Shinichiro Toya

发明人： Keishi Nakayama ,  Keitaro Uehara ,  Takashi Aoyagi ,  Shinichiro Toya

IPC分类号： G06F13/20 ,  G06F3/00 ,  G06F13/36 ,  G06F13/00

CPC分类号： G06F13/4022

摘要： To make it possible to take over an IO configuration that is assigned to logical partitions in reallocation of the logical partitions, and to make an IO access work normally. A computer system has a server having an IO bridge, a switch that has a first IO bridge for connecting with the IO bridge of the server through a bus and plural second IO bridges for connecting to plural IO devices through a bus, and bus number assignment management means for fixedly assigning mutually different PCI bus numbers to the plural second IO bridges.

摘要翻译： 为了能够接管分配给逻辑分区的IO配置，重新分配逻辑分区，并使IO访问正常工作。 计算机系统具有服务器，其具有IO桥，交换机具有用于通过总线与服务器的IO桥连接的第一IO桥，以及用于通过总线连接到多个IO设备的多个第二IO桥，以及总线号分配 用于将相互不同的PCI总线号固定地分配给多个第二IO桥的管理装置。

公开(公告)号：US08751637B2

公开(公告)日：2014-06-10

申请号：US12980606

申请日：2010-12-29

申请人： Hideki Okita ,  Masahiro Yoshizawa ,  Keitaro Uehara

发明人： Hideki Okita ,  Masahiro Yoshizawa ,  Keitaro Uehara

IPC分类号： G06F15/173 ,  H04L12/24 ,  H04L12/26

CPC分类号： H04L43/0817 ,  H04L41/142

摘要： A failure factor can be promptly carved up by effectively prioritizing the ports to which MEPs are assigned in a condition where the maximum number of MEPs is limited. A network monitoring server monitors a network having plural switch devices. The plural switch devices each have a communication confirmation function using transmission and reception of a monitoring frame between ports. An upper limit is set to the number of ports to which maintenance end points (MEPs) are assignable among the respective ports of the plural switch devices. The network monitoring server calculates, for each of the plural switch devices, a monitoring priority corresponding to the probability of failures of a device connected to a network path including the port or the port, and generates data for displaying the calculated monitoring priority of each port in association with each port.

摘要翻译： 通过在限制MEP最大数量的条件下有效优先分配MEP分配的端口，可以及时刻录故障因素。 网络监视服务器监视具有多个交换设备的网络。 多个开关装置各自具有使用端口之间的监视帧的发送和接收的通信确认功能。 对多个开关装置的各个端口中的维护端点（MEP）可分配的端口数设定上限。 所述网络监视服务器针对所述多个交换机装置中的每一个计算与连接到包括所述端口或所述端口的网络路径的设备的故障概率相对应的监视优先级，并且生成用于显示所计算的每个端口的监视优先级的数据 与每个港口相关联。

公开(公告)号：US20100036995A1

公开(公告)日：2010-02-11

申请号：US12486927

申请日：2009-06-18

申请人： Keishi Nakayama ,  Keitaro Uehara ,  Takashi Aoyagi ,  Shinichiro Toya

发明人： Keishi Nakayama ,  Keitaro Uehara ,  Takashi Aoyagi ,  Shinichiro Toya

IPC分类号： G06F13/00

CPC分类号： G06F13/4022

摘要： To make it possible to take over an IO configuration that is assigned to logical partitions in reallocation of the logical partitions, and to make an IO access work normally. A computer system has a server having an IO bridge, a switch that has a first IO bridge for connecting with the IO bridge of the server through a bus and plural second IO bridges for connecting to plural IO devices through a bus, and bus number assignment management means for fixedly assigning mutually different PCI bus numbers to the plural second IO bridges.

摘要翻译： 为了能够接管分配给逻辑分区的IO配置，重新分配逻辑分区，并使IO访问正常工作。 计算机系统具有服务器，其具有IO桥，交换机具有用于通过总线与服务器的IO桥连接的第一IO桥，以及用于通过总线连接到多个IO设备的多个第二IO桥，以及总线号分配 用于将相互不同的PCI总线号固定地分配给多个第二IO桥的管理装置。