公开(公告)号：US20110268113A1

公开(公告)日：2011-11-03

申请号：US13067908

申请日：2011-07-06

申请人： Jun Suzuki ,  Masanori Takashima

发明人： Jun Suzuki ,  Masanori Takashima

IPC分类号： H04L12/56

CPC分类号： G06F9/4856 ,  G06F9/54

摘要： A packet communication system includes a virtual machine unit; an origin-of-movement communication unit; a movement-destination communication unit; and a storage unit which stores a packet destined for the virtual machine unit while the virtual machine unit moves from the origin-of-movement communication unit to the movement-destination communication unit and transfers the stored packet to the movement-destination communication unit.

摘要翻译： 分组通信系统包括虚拟机单元; 运动起源通信单元; 移动目的地通信单元; 以及存储单元，其在虚拟机单元从移动目的地通信单元移动到移动目的地通信单元的同时存储去往虚拟机单元的分组，并将存储的分组传送到移动目的地通信单元。

公开(公告)号：US10554585B2

公开(公告)日：2020-02-04

申请号：US14122003

申请日：2012-05-29

申请人： Mamoru Takajo ,  Masanori Takashima

发明人： Mamoru Takajo ,  Masanori Takashima

IPC分类号： H04L12/947

摘要： On the OpenFlow network/normal network boundary, the route cannot be appropriately changed under a communication trouble with the controller. In this invention, the controller makes the route formed by internal switches optimal by updating the Flow table of each internal switch. Among internal switches, the internal switch A forms the optimal route with an external switch to become an active internal switch. The internal switch B forms a redundant route with the external switch to become a standby internal switch. When inability of the connection between the controller and the internal switch A is detected, the internal switch A performs a link-down of the port for connecting to the external switch. When the link-down of the port connecting to the internal switch A is detected, the external switch changes the route so that the traffic to the active internal switch A is transferred to the standby switch B.

公开(公告)号：US20150003291A1

公开(公告)日：2015-01-01

申请号：US14377848

申请日：2013-03-18

申请人： Seiji Oikawa ,  Masanori Takashima

发明人： Seiji Oikawa ,  Masanori Takashima

IPC分类号： H04L29/12 ,  H04L12/24

CPC分类号： H04L61/35 ,  H04L12/4625 ,  H04L41/12 ,  H04L45/02 ,  H04L45/42 ,  H04L61/6022

摘要： In a network including: a first node(s) controlled by a control apparatus in a centralized manner; a second node(s) storing, for a predetermined period, entries in each of which, based on a source address of a received packet, its own port and information about a node address(es) to which packets can be forwarded from the port are associated with each other and performing packet forwarding by referring to a group of entries and determining a port corresponding to a packet destination; and the control apparatus, the control apparatus causes the first node(s) connected to the second node(s) to transmit a destination learning packet to the second node(s) to cause the second node(s) to learn information about an address(es) to which packets can be forwarded from a port(s) of the second node(s) at predetermined time intervals.

摘要翻译： 一种网络，包括：由集中式的控制装置控制的第一节点; 存储预定时间段的第二节点，其中每个节点基于接收到的分组的源地址，其自己的端口和关于可以从该端口向其转发分组的节点地址的信息 通过参考一组条目并确定与分组目的地对应的端口来进行分组转发; 和所述控制装置，所述控制装置使连接到所述第二节点的所述第一节点向所述第二节点传送目的地学习分组，以使所述第二节点学习关于地址的信息 （es），可以以预定的时间间隔从第二节点的端口转发分组。

公开(公告)号：US20130046882A1

公开(公告)日：2013-02-21

申请号：US13520773

申请日：2011-01-04

申请人： Masanori Takashima ,  Youichi Hidaka ,  Tetsu Izawa

发明人： Masanori Takashima ,  Youichi Hidaka ,  Tetsu Izawa

IPC分类号： G06F15/173

CPC分类号： H04L45/42 ,  H04L45/18 ,  H04L45/38 ,  H04L45/48 ,  H04L45/54

摘要： In the case of a system where a communication device is remote-controlled from a control server, depending on timing when an entry is written into a forwarding table of the communication device from the control server, there is a possibility that a loop and discard of the traffic occur. Therefore, for registration, rewriting and deletion of the entry of the forwarding table which are performed on the communication device 102 by the control server 101; the communication device 102 verifies contents of control instructions from the control server 101; the communication device 102 holds the result of this verification; the control server 101 acquires the result of the verification; the control server 101 judges whether or not a series of controls can be executed based on the results of the verification from the plurality of communication devices 102; and then the control server 101 invokes the execution at the plurality of communication devices 102.

摘要翻译： 在从控制服务器远程控制通信装置的系统的情况下，根据从控制服务器将条目写入到通信装置的转发表中的定时，存在循环和丢弃 交通发生。 因此，用于登记，重写和删除由控制服务器101在通信设备102上执行的转发表的条目; 通信设备102从控制服务器101验证控制指令的内容; 通信装置102保持该验证的结果; 控制服务器101获取验证结果; 控制服务器101基于来自多个通信装置102的验证结果判断是否能够执行一系列控制; 然后控制服务器101调用多个通信设备102的执行。

公开(公告)号：US20110289235A1

公开(公告)日：2011-11-24

申请号：US13137251

申请日：2011-08-01

申请人： Masanori Takashima

发明人： Masanori Takashima

IPC分类号： G06F15/173

CPC分类号： H04L67/1002 ,  H04L67/2857

摘要： A distributed processing system 100 includes: a plurality of application servers 1031 to 103n that execute a plurality of applications 1321 to 132n and detect a communication state between a client 105 and the applications 1321 to 132n; a load distribution control server 101 that decides a communication path between the client 105 and the applications 1321 to 132n on a basis of the communication state detected by the application servers 1031 to 103n to perform load distribution; and an intermediate communication apparatus 102 that transfers flow data between the client 105 and the application servers 1031 to 103n.

摘要翻译： 分布式处理系统100包括：执行多个应用1321〜132n的多个应用服务器1031〜103n，检测客户端105与应用1321〜132n之间的通信状态; 基于由应用服务器1031〜103n检测到的通信状态来决定客户机105和应用1321〜132n之间的通信路径以进行负载分配的负载分配控制服务器101; 以及在客户端105和应用服务器1031〜103n之间传送流数据的中间通信装置102。

公开(公告)号：US09787580B2

公开(公告)日：2017-10-10

申请号：US13995921

申请日：2011-12-27

申请人： Masanori Takashima ,  Tomohiro Kase

发明人： Masanori Takashima ,  Tomohiro Kase

IPC分类号： H04L12/721 ,  H04L12/931 ,  H04L12/725 ,  H04L12/781 ,  H04L12/715 ,  H04L12/741 ,  H04L12/851

CPC分类号： H04L45/38 ,  H04L45/306 ,  H04L45/52 ,  H04L45/64 ,  H04L45/74 ,  H04L47/2483 ,  H04L49/70

摘要： In an OpenFlow network, a “proactive type” is attained and hardware (HW) performance problem is solved. Specifically, in the OpenFlow network, each of a plurality of switches executes, on a reception packet that meets a rule of an entry registered in its own flow table, an operation based on an action defined in the entry. A controller registers an entry, in which an identifier unique to a path calculated based on a physical topology of a network composed of the plurality of switches is set as a rule and an output from a predetermined output port as an action, in each of the plurality of switches before communication is started among the plurality of switches.

公开(公告)号：US09215175B2

公开(公告)日：2015-12-15

申请号：US13821926

申请日：2011-09-05

申请人： Masanori Takashima ,  Tomohiro Kase ,  Hiroshi Ueno ,  Takahisa Masuda ,  Suhun Yun

发明人： Masanori Takashima ,  Tomohiro Kase ,  Hiroshi Ueno ,  Takahisa Masuda ,  Suhun Yun

IPC分类号： H04L12/28 ,  H04L12/801 ,  H04L12/717 ,  H04L12/46 ,  H04J1/16 ,  H04L12/721 ,  H04L12/707

CPC分类号： H04L47/12 ,  H04L12/462 ,  H04L45/24 ,  H04L45/38 ,  H04L45/42 ,  H04L45/66 ,  H04L61/103

摘要： A computer system of the present invention includes a controller which sets a flow entry to each of a plurality of switches before a packet is forwarded from a plurality of nodes, and each of switches which transfers a reception packet which contains a destination address defined in the flow entry, to a destination node defined in the flow entry regardless of a transmission source address of the reception packet.

摘要翻译： 本发明的计算机系统包括控制器，其在从多个节点转发分组之前，将多个交换机中的每一个设置流入口，并且每个交换机传送包含在 流入口到达流入口中定义的目的地节点，而不管接收分组的发送源地址如何。

公开(公告)号：US20130279371A1

公开(公告)日：2013-10-24

申请号：US13995921

申请日：2011-12-27

申请人： Masanori Takashima ,  Tomohiro Kase

发明人： Masanori Takashima ,  Tomohiro Kase

IPC分类号： H04L12/56

CPC分类号： H04L45/38 ,  H04L45/306 ,  H04L45/52 ,  H04L45/64 ,  H04L45/74 ,  H04L47/2483 ,  H04L49/70

摘要： In an OpenFlow network, a “proactive type” is attained and hardware (HW) performance problem is solved. Specifically, in the OpenFlow network, each of a plurality of switches executes, on a reception packet that meets a rule of an entry registered in its own flow table, an operation based on an action defined in the entry. A controller registers an entry, in which an identifier unique to a path calculated based on a physical topology of a network composed of the plurality of switches is set as a rule and an output from a predetermined output port as an action, in each of the plurality of switches before communication is started among the plurality of switches.

摘要翻译： 在OpenFlow网络中，实现了“主动型”，并解决了硬件（HW）性能问题。 具体地，在OpenFlow网络中，多个交换机中的每一个在满足在其自己的流程表中登记的条目的规则的接收分组上执行基于条目中定义的动作的操作。 控制器登记条目，其中将基于由多个交换机组成的网络的物理拓扑计算出的路径唯一的标识符作为规则设置为规则，并将来自预定输出端口的输出作为动作进行设置， 在多个开关之间开始通信之前的多个开关。

公开(公告)号：US20130086295A1

公开(公告)日：2013-04-04

申请号：US13580620

申请日：2011-02-18

申请人： Youichi Hidaka ,  Masashi Hayashi ,  Shihomi Sato ,  Tsugio Okamoto ,  Takashi Yokota ,  Masanori Takashima ,  Koichi Tsuchiya ,  Minaxay Phila Vongh ,  Tetsu Izawa

发明人： Youichi Hidaka ,  Masashi Hayashi ,  Shihomi Sato ,  Tsugio Okamoto ,  Takashi Yokota ,  Masanori Takashima ,  Koichi Tsuchiya ,  Minaxay Phila Vongh ,  Tetsu Izawa

IPC分类号： G06F13/40

CPC分类号： G06F13/4022 ,  H04L41/00 ,  H04L49/35

摘要： In a switching node, high-speed and advanced service protocol processing function is achieved by utilizing an external control server without affecting performance of conventional service protocol processing. Specifically, a forwarding engine has PCI express and an LAN interface. Depending on a type of an input packet, destination of the packet is switched to the PCI express side for conventional network service and to the LAN interface side for extended network service that cooperates with the external control server. A CPU having the PCI express and the LAN interface is provided ahead of the LAN interface. The CPU performs communication of service inquiry with the external control server at high speed via the LAN interface. After response from the control server is obtained, setting of the forwarding engine is performed through the PCI express.

摘要翻译： 在交换节点中，通过利用外部控制服务器，不影响传统业务协议处理的性能，实现了高速，高级的业务协议处理功能。 具体来说，转发引擎具有PCI Express和LAN接口。 根据输入分组的类型，分组的目的地被切换到用于常规网络服务的PCI Express侧和用于与外部控制服务器协作的扩展网络服务的LAN接口侧。 具有PCI Express和LAN接口的CPU在LAN接口之前提供。 CPU通过LAN接口与外部控制服务器高速进行业务查询通信。 获得控制服务器的响应后，通过PCI express进行转发引擎的设置。

公开(公告)号：US20130077481A1

公开(公告)日：2013-03-28

申请号：US13520776

申请日：2011-01-04

申请人： Minaxay Philavong ,  Masanori Takashima ,  Youichi Hidaka ,  Tetsu Izawa

发明人： Minaxay Philavong ,  Masanori Takashima ,  Youichi Hidaka ,  Tetsu Izawa

IPC分类号： H04L12/56

CPC分类号： H04L49/25 ,  H04L41/0668 ,  H04L45/22 ,  H04L45/38 ,  H04L45/66

摘要： In a network system in which a switch forwarding packets and a control server determining route information are separated, it is desired to achieve a redundancy of the control channel by an out-of-band control channel and an in-band control channel, when the switch receives a control message regarding the flow entry registration and the like based on the route information from the control server. Specifically, the separated switch and control server are connected by a control channel for sending and receiving the control message. The switch is not only connected to a control server via an out-of-band control channel by a route dedicated to the control message, but also connected to another control server via an in-band control channel by a route which is common with a normal data communication for determining the route information, to achieve a redundancy of the control channel.

摘要翻译： 在将转发分组的交换机转发分组和确定路由信息的控制服务器分离的网络系统中，期望通过带外控制信道和带内控制信道来实现控制信道的冗余，当 交换机基于来自控制服务器的路由信息​​接收关于流入口登记等的控制消息。 具体地，分离的开关和控制服务器通过用于发送和接收控制消息的控制信道连接。 交换机不仅通过带外控制信道通过专用于控制消息的路由连接到控制服务器，而且还经由带内控制信道通过与 用于确定路由信息的正常数据通信，以实现控制信道的冗余。