• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A server (10) of a management system of an interconnection network, the server comprising - a data structure (2) associative key-value configured to store a global state of the interconnection network; - the following connection interfaces (3-6) ○ a configuration server interface (3) configured to communicate, in response to a configuration request to join the management system, the configuration of said management network; A publication server interface (4) configured to broadcast an update in the key-value associative data structure (2); ○ a collection server interface (5) configured to update, in response to an update request, the associative key-value data structure (2); A snapshot server interface (6) configured to communicate, in response to a snapshot request, a snapshot of a subset of the global state.
    公开号:FR3022420A1
    申请号:FR1455446
    申请日:2014-06-13
    公开日:2015-12-18
    发明作者:Pierre Vigneras；Sebastien Miquee
    申请人:Bull SA；
    IPC主号:
    专利说明:

    [0001] The invention relates to a management system for managing an interconnection network.
    [0002] Here, the term "interconnection network" refers to any dedicated computing network (such as an InfiniBand network), or more generally any collection of computing elements, in particular distributed processors, with physical communication links between them. . A management system, using a secondary network, which may be of the Ethernet type, is configured to manage this computing network in an out-of-band mode. However, with the growth in the size of supercomputers, the topologies of high-performance computing networks are becoming more complex and denser. It follows that the out-of-band management, by means of a dedicated management network also called "secondary network", of the interconnection network of a supercomputer requires more than one component to: react to events that could occur in this interconnection network (cable, port, switch, or node failures, for example); check for any changes in the topology related to changes in cables or switches (connectivity, compliance with theoretical topologies); calculate the new routing tables based on the changes and updates loaded into the switches. In addition, for reasons of scalability and robustness, each of the above tasks is to be performed by separate processes, potentially from different dedicated machines. To this end, an effective communication mechanism allows these components to share a common global state by interacting. This mechanism must be ensured by the management system. The messages are exchanged on the secondary network (or management network) of the supercomputer interconnection network. The management system enables the processes in charge of managing the interconnection network, to communicate and to share a global state, a subset of which represents the state of the supercomputer's interconnection network (ie the whole network). the status of the equipment that makes up the supercomputer interconnection network). In order to ensure communication between processes running on machines different from the management network, this management network must, in particular, be able to support a large number of processes (several hundreds or even thousands); allow a process to join or leave the management system at any time (due to a downtime, or update for example); maintain an up-to-date global status of the supercomputer's interconnection network; maintain a coherent global state ("eventual consistency" in English terminology) of the interconnection network of the supercomputer; represent this global state in the form of a "key-value" type table; share this global state between the different interconnected processes of the management system; allow at any time and any of the management system processes to update a subset or even the entire state of the interconnection network; manage the frequency of updates of the global state (preferably, at most a few hundred per second); restricting access to the global state, with some processes having to / want to see only a subset of the global state; be able to keep the global state in RAM even though it can have a relatively large number of inputs (usually several hundreds of thousands). In this regard, there are state-sharing systems such as distributed hash tables. However, none of the existing solutions satisfies all the above requirements that a management system must meet In the present case, a distributed hash table can not offer the last two constraints stated above. An object of the present invention is to provide an interprocess communication mechanism in the form of a management system meeting the above requirements.
    [0003] Another project of the present invention is to provide a client / server type communication architecture for interconnecting distributed processes. Another object of the present invention is to provide a management system for an interconnection network based on inter-process communication. Another object of the present invention is to provide an asynchronous and disconnected interprocess communication mechanism. Another project of the present invention is to propose a method for managing the interconnection network of a supercomputer. For these purposes, the invention relates, in a first aspect, to a server of a management system of an interconnection network, the server comprising: a key-value associative data structure configured to store a state overall interconnection network; the following connection interfaces o a configuration server interface configured to communicate, in response to a configuration request to join the management system, the configuration of said management system; o a publishing server interface configured to broadcast an update to the key-value associative data structure; o a collection server interface configured to update, in response to an update request, the associative key-value data structure; o A snapshot server interface configured to communicate, in response to a snapshot request, a snapshot of a subset or the entire global state. The server of an interconnect network management system has, according to various implementations, the following characters, if any combined: the configuration server interface is configured to communicate the address of the publisher server interface , the address of the collection server interface, and the address of the snapshot server interface; the associative key-value data structure is a hash table; the implementation of this server is based on the connection server interface event paradigm, this connection server interface being the configuration server interface, the collection server interface or the snapshot server interface. The invention relates, in a second aspect, to a client of a management system of an interconnection network, this client comprising a client process; a snapshot client interface configured to retrieve a snapshot of a subset or the entire global status of the interconnect network; a subscription client interface configured to receive an update of the overall state of the interconnect network; a configuration client interface configured to retrieve the management system configuration; a publishing client interface so that the client process can publish an update of the overall state of the interconnect network. The client of an interconnection network management system further comprises a business process, which business process is provided with a publishing interface so that the business process can publish an update of the business process. overall state of the interconnection network. Advantageously, the data published by the client process is a message based on the "key-value" paradigm. The invention relates, in a third aspect, to a management system for managing an interconnection network comprising the server and the client presented above. According to a fourth aspect, the invention relates to a supercomputer comprising an interconnection network and the management system mentioned above. The supercomputer further comprises: - a topology manager configured to check the topology of the interconnection network; a supervisory module configured to retrieve an alarm sent by a network equipment of the interconnection network, and to update the management system accordingly; a routing calculator configured to calculate a routing table for data received from the management system. Other objects and advantages of the invention will emerge in the light of the description of the embodiments, given below with reference to the appended drawings, in which: FIG. 1 illustrates a server of a management system according to a production ; Figure 2 illustrates a client of a management system according to one embodiment; FIG. 3 illustrates the implementation of a management system according to one embodiment.
    [0004] The management system of an interconnection network is based on a client / server type network architecture implementing different communication paradigms as a function of the connection interfaces. Referring to Figure 1, the server side (or server implementation) of this management system network architecture includes a server 10 of the management network. The server 10 is configured to hold the overall state of the interconnect network. This global state includes information about the supercomputer interconnect network. This global state is stored in a key-value associative data structure. This data structure 2 is stored in a random access memory of the server 10. The associative data structure 2 key-value is a data container, preferably local to the server 10, which has a particular protocol for adding, removing and the search for elements. This associative key-value data structure 2 associates a key with a value. The uniqueness of the keys must be guaranteed by the issuing processes. If an update is performed on an already existing key, the old value is overwritten by the new one. In one embodiment, the associative key-value data structure 2 is an associative table, also called a hashmap or hashmap, having a predefined association or hash function. Advantageously, this particular data structure allows quick access to a value depending on a key. The server 10 of the management network further comprises a configuration server interface 3 configured to manage the configuration request (s) from a client wishing to join the management system; a publisher server interface 4 configured to broadcast, to the clients connected to the management system, updates taking place in the associative key-value data structure 2, an input of the key-value associative data structure 2; a collection server interface 5 configured to handle update requests (updates, typically in the form of a key-value message) from the client processes connected to the management system; a snapshot server interface 6 (or "snapshot" in English terminology) configured to send the current state, that is to say a snapshot of all or only a subset of the associative key-value data structure 2 (the filtering can be done at server level 10, in order to limit the use of bandwidth), in response to a snapshot request from a client process connected to the system of management.
    [0005] The configuration server interface 3 makes it possible to communicate the configuration of the management system to a client wishing to join this management system of the interconnection network. The sent configuration comprises, in particular, the addresses (or URLs for "Uniform Resource Locator") of the other connection interfaces mentioned above, namely the publication server interface 4, the collection server interface 5, and the snapshot server interface 6. Preferably, the 3-6 connection server interfaces are ZeroMQ sockets (the document at the following link: http://zeromq.org/intro:read-the-manual specifies ZeroMQ or ZMQ sockets ). This embodiment makes it possible to avoid the problems of the request / response paradigm. It should be noted that in the server implementation, shown in FIG. 1, of the management system, the configuration server interface 3 does not interact with the key-value associative data structure 2; the collection server interface 5 interacts with this key-value associative data structure 2 in write mode to store key-value inputs therein; the publisher server interface 4 and the snapshot server interface 6 interact with the associative key-value data structure 2 in read mode. This server implementation of the management system can be done in C, Python, or any other appropriate programming language, as well as in a single-threaded or multi-threaded mode. In a single-threaded implementation of Server 10, the server executes the following loop, based on the Connection Server Interface Event Paradigm 3.5, 6, WHILE true according to "Connection Server Interface" CAS "Configuration Server Interface »Send the configuration data through the CAS configuration interface" collection server interface "send the update through the publisher interface update the data structure 2 CAS" snapshot server interface "send a instantaneous via the CAS snapshot server interface "the delay of a timer has expired" send a FIN FINANCE message Advantageously, this implementation based on the connection server interface event paradigm 3, 5, 6 can react quickly to an action from a client process (query or update). The conditional statement "SELON" manages the three connection server interfaces (configuration 3, collection 5 and snapshot 6) and a predefined waiting time (timeout). It returns a list including the connection interface or interfaces triggered by an event of a connection server interface 3, 5, 6. If no event is produced while the delay time has expired, the server 10 sends to management system clients a presence message (designated by "Heartbeat message" in English terminology). The particular order in which the connection server interfaces 3, 5, 6 are processed makes it possible, in the first place, for a client to connect to the server 10 and to configure itself using the addresses (URLs) of the other server interfaces of 4-6 connection communicated to it via the configuration server interface 3; second, if the client automatically sends (or, in English terminology, "push") an update, to take into account this update as soon as possible by the server 10 and publish it to all the customers (including the one who caused this update), this update may trigger actions at other management system clients; third, a request for a snapshot (in English terminology, a "snapshot") to be processed last because it usually takes the longest time to execute.
    [0006] Advantageously, in a single-threaded implementation of the server 10, a locking mechanism to prevent simultaneous and concurrent access to the data structure 2 and / or the simultaneous use of a connection server interface 3-6 is not necessary. The management system server 10 receives and processes key-value entries that can be interpreted as commands. These entries may be from a management system client. Advantageously, this makes it possible to have hidden entries, especially when the data structure 2 is written in system files (the case of a system based on Unix for example).
    [0007] Figure 2 illustrates a client implementation or the "client side" of the client-server architecture of the management system of an interconnection network. Here, a client 20 of the management system means a client process 21 which is generally configured to support at least one business process 31 for the execution of a business code 30. The client 20 of the management system an interconnect network, includes a snapshot client interface 26 configured to retrieve a snapshot of the overall state of the interconnect network, or a subset thereof; a subscription client interface 24 configured to receive updates from the management system server; a configuration client interface 23 configured to retrieve the configuration of the management system server, in particular the addresses (URLs) of the server connection server interfaces; a plurality of publishing client interfaces 25 so that the client process 21 and any other business process 31 of a business code 30 can publish, by means of "key-value" entities, updates of (or, more generally, data about their respective states via an appropriate publishing client interface. It should be noted that a business code 30 may include, for its own purposes, more than one business process 31 to which a publishing client interface 25 is associated; a communication client interface 32 between the business process 31 of the business code 30 and the client process 21. The client process 21 manages the incoming communications of the client 20 of the management system. For this, this client process 21 has its own input client interfaces 24 and 26, that is to say that of snapshot 26 and that of subscription 24.
    [0008] The publishing client interface 25 allows a business process 31 to update the overall state of the interconnect network that is shared and stored on the server side. In Figure 3, the client 20 is connected to the server 10 of the management system. In order to connect to the server 10, the client 20 first connects to the configuration server interface 3 to receive all the configuration parameters of the management system, including the addresses (URLs) of the other interfaces. server server connection 4-6 10. In one embodiment, upon receipt on the configuration server interface 3 of a configuration request (for example, of type CONFIG ), the server 10 (see the link between the connection interfaces 3 and 23) communicates via the configuration server interface 3, in response to this request, the addresses (URLs) of the other three connection interfaces 4-6; and sends, via this configuration server interface 3, a message indicating the end of the configuration step (of type, for example Fin_CONFIG).
    [0009] Upon receipt of the configuration parameters, the client 20 decides to retrieve or not a snapshot (a "snapshot") of the current global state of the interconnection network (see the connection between the connection interfaces 6 and 26). If so, the client requires, from the snapshot server interface 6, a snapshot of the overall state of the interconnect network that is stored in the associative key-value data structure 2; the server 10 waits for the client 20 to receive the hierarchies to which it wishes to subscribe (these hierarchies may, in particular, be different from those specified during a connection to the publication server interface 4 of the server 10).
    [0010] Then, the server 10 sends to the client 20, using the snapshot server interface 6, a snapshot of the state of the required hierarchies; the server 10 sends the current revision number of the key-value associative data structure 2. It follows that it is the responsibility of the client 20 to connect to the publisher interface 4 (see the connection between the connection interfaces 4 and 24) to receive updates, and the interface collection server 5 (see the connection between the connection interfaces 5 and 25) to possibly send updates (or new entries). Preferably, the updates published by the server 10 are time stamped. Advantageously, the time stamping (or "timestamping" in English) of the server side ensuring the consistency ("eventual consistency") of the overall state within each client 20. It should be noted that a client 20 can Ignore updates sent during snapshot recovery when they are earlier (thanks to the timestamp) than the snapshot received. Apart from the connection time of a client 20, the server 10 is configured to respond to each update request and to send, regularly, heartbeat messages. On receipt of an update request, or more generally of an order, via the collection server interface 5, if this command does not concern the server 10 or if it is not a special command ( such as a shutdown command), it is transferred to other clients via the Publisher 4 interface; the server 10 applies to the data structure 2 the content of this update request. In the case of a server control message, this message is first compared to the "empty" command which requires the deletion of all the contents of the data structure 2. Using the special fields of the data of input, such as "empty" or "ttl" for "time to live", the server 10 can delete this input data from the local data structure 2. Otherwise, if no input data exists, then it is added or updated in the data structure 2. As examples of requests that the server 10 can receive, quote .bbctrILserver_kill_: require the server 10 to leave (this message not being broadcast by the server 10); .bbctrli_clear_all_: require the server 10 and all the clients to delete all the contents of the data structure (this message being broadcast by the server 10); .bbctrli_hugz_: a presence message from the server.
    [0011] The basic entity here is a message based on the "key-value" paradigm. This entity provides, in addition to the key-value pair, other information such as the sequence number, the unique universal identifier of this entity, the identity of its issuer, or the lifetime of the entity ( "Time to Live").
    [0012] This information can be sent in one or more data frames. Preferably, this information is sent in a first frame containing the key for the subscription mechanism, and a second frame containing the rest of the information (data, identifier of the sender, sequence number for example). Preferably, a dynamic polling time is calculated before the pollster of the collection server interface 5 initiates a period of inactivity.
    [0013] Advantageously, the server implementation 10 of the management network makes it possible to supply the configuration data to the clients; provide snapshots of the overall state of the interconnect network to the newly connected management system clients; and to transfer updates in the global state of the interconnect network to all clients connected to the management system.
    [0014] In one embodiment, the management system is used for managing the interconnection (or calculation) network of a supercomputer. For this, the following modules may be required a topology manager responsible for verifying the topology (the equipment status for example or the equipment connection diagram) of the interconnection network; a supervision module responsible for retrieving the alarms sent by the various network equipment of the supercomputer (in particular the switches), correlating them in order to update the management system with relevant data (status updates for example) ; a routing calculator responsible for calculating the routing tables for data received from the management system. When the additional components above start, the server 10 of the management network is already present so that all the data published by the topology manager can be retrieved by the other clients of the server 10. Advantageously, status days are made unitarily, and after aggregation, a special key, called trigger, is sent. It is at the reception of the latter that the routing calculator triggers its phase of calculating the new routing tables. Thus, it is possible to differentiate easily and quickly the different types of messages sent by the management system server - updates of the overall state of the interconnection network; - the triggers that trigger an action. This results, advantageously, an aggregation of events to gain efficiency. In particular, the routing calculator does not calculate the routing tables for each status change of a single device. In this respect, a key prefix mechanism not only allows each client to subscribe to a subset of the global state, but also to make this differentiation. In addition, this mechanism makes it possible to add to the management system other types of information that are found in specific subsets. By way of example, the following prefixes can be used in a management system: @ interconnect /: set of entries concerning the status of the equipment of the interconnection network; triggers /: set of inputs corresponding to triggers; / mngt /: set of entries containing connection information to the various devices of the interconnection network; .bbctr1 /: set of entries to control the management system, both client and server side (eg a command to erase the entire data table). Advantageously, the various embodiments described above use different communication paradigms as a function of the connection interfaces. For example, the server interface 30 of publication 4 of the server 10 operates in "broadcast" mode and can use a "multicast" protocol (symmetrical with the subscription server interface 24 on the client side 20). In addition, a client 20 can subscribe to one or more key prefixes, which allows it to retrieve only a subset of the global state stored in the data structure 2.
    权利要求:
    Claims (10)
    [0001]
    REVENDICATIONS1. A server (10) of an interconnection network management system, which server comprises a key-value associative data structure (2) configured to store a global state of the interconnection network; the following connection interfaces (3-6) o a configuration server interface (3) configured to communicate, in response to a configuration request to join the management system, the configuration of said management system; an advertisement server interface (4) configured to broadcast an update in the associative key-value data structure (2); o a collection server interface (5) configured to update, in response to an update request, the associative key-value data structure (2); o A snapshot server interface (6) configured to communicate, in response to a snapshot request, a snapshot of a subset of the global state.
    [0002]
    The server (10) of claim 1, wherein the configuration server interface (3) is configured to communicate the address of the publisher server interface (4), the address of the server interface of collection (5), and the address of the snapshot server interface (6).
    [0003]
    The server (10) of claim 1 or 2, wherein the associative key-value data structure (2) is a hash table.
    [0004]
    The server of any one of the preceding claims, wherein the implementation of this server is based on the connection server interface event paradigm, which connection server interface is the configuration server interface (3), the collection serverinterfaces (5) or the snapshot server interface (6).
    [0005]
    5. A client (20) of an interconnection network management system, which client comprises a client process (21); a snapshot client interface (26) configured to retrieve a snapshot of a subset of the overall state of the interconnect network; a subscription client interface (24) configured to receive an update of the overall state of the interconnect network; a configuration client interface (23) configured to retrieve the configuration of the management system; a publishing client interface (25) so that the client process (21) can publish an update of the overall state of the interconnect network.
    [0006]
    The client of the preceding claim, further comprising a business process, wherein the business process is provided with a publishing interface (25) so that the business process can publish an update of the overall state of the network. interconnection.
    [0007]
    The client of claim 5 or 6, wherein the published update is a message based on the "key-value" paradigm.
    [0008]
    A management system for managing an interconnection network comprising a server according to any one of claims 1 to 4 and a client according to any one of claims 5 to 7.
    [0009]
    9. A supercomputer comprising an interconnection network and a management system of said interconnection network according to claim 8.
    [0010]
    The supercomputer of the preceding claim further comprising: - a topology manager configured to check the topology of the interconnection network; a supervisory module configured to recover an alarm sent by a network equipment of the interconnection network, and to update the management system accordingly; a routing calculator configured to calculate a routing table for data received from the management system.
    类似技术:
    公开号 | 公开日 | 专利标题
    EP2955875B1|2018-10-10|Server, client and system for managing an interconnection network
    US8024477B2|2011-09-20|System and method to establish a peer-to-peer IT backbone
    AU2014215935B2|2015-12-10|A system and method for managing message queues in a peer-to-peer communication network
    WO2006059040A1|2006-06-08|Method and system for the dynamic adaptation of service quality metrics in an ad hoc network
    EP2791798B1|2020-09-30|Software bus
    FR2948247A1|2011-01-21|METHOD AND SYSTEM FOR HIGH PERFORMANCE AND AUTOMATED MANAGEMENT OF VIRTUAL NETWORKS.
    WO2008012273A1|2008-01-31|Method of synchronization between a mobile equipment unit and a smart card
    EP2087693A2|2009-08-12|Method for reacting to the broadcast of a file in a p2p network
    EP1652346B1|2009-08-12|Method of locating mobile communicating objects within a communications network
    FR2969890A1|2012-06-29|METHOD AND DEVICE FOR COMMUNICATING DIGITAL DATA
    EP3050272B1|2017-08-02|Device and method for managing subscriptions to point-to-multipoint transmissions
    FR3103664A1|2021-05-28|Distributed storage system for storing contextual data
    FR2919140A1|2009-01-23|METHOD FOR EXCHANGING MESSAGES BETWEEN SESSION DATA SERVER AND CLIENT SERVICES
    FR2816419A1|2002-05-10|Method for dividing load between servers on distributed system, comprises determination of tasks for transfer by a first server followed by negotiation with load control units on other servers
    Lin et al.2021|A Programmable and Reliable Publish/Subscribe System for Multi-Tier IoT
    EP3949457A1|2022-02-09|Message transmission in a multi-terminal context
    EP3563558A1|2019-11-06|Computer network of nodes communicating with one another by peer-to-peer messages and associated method for interconnecting between nodes
    WO2013088019A1|2013-06-20|Method and computer program for managing multiple faults in a computing infrastructure comprising high-availability equipment
    FR3067832A1|2018-12-21|SUPPLY OF INTER-GROUP SERVICES
    FR3081238A1|2019-11-22|DATABASE BASE CALLING
    EP2341427B1|2012-08-22|Autonomous restart of peer-to-peer network nodes
    FR2969448A1|2012-06-22|Method for optimizing transmission of message to application server that receives messages to client terminal for events notification on e.g. smart phone, involves determining remaining retention period according to maximum retention period
    WO2005008492A2|2005-01-27|System for the distributed management of computer resources and calculations
    Baehni et al.2006|Community-Aware Event Dissemination
    FR3031204A1|2016-07-01|METHOD AND SYSTEM FOR ALLOCATING PROCESSING RESOURCES FOR EXECUTING A PROGRAM OF CUSTOMER EQUIPMENT
    同族专利:
    公开号 | 公开日
    US20150365284A1|2015-12-17|
    FR3022420B1|2018-03-23|
    EP2955875B1|2018-10-10|
    US9866437B2|2018-01-09|
    JP6304499B2|2018-04-04|
    EP2955875A1|2015-12-16|
    JP2016005275A|2016-01-12|
    ES2700929T3|2019-02-20|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    US20030041094A1|1998-05-29|2003-02-27|Marco Lara|Web server content replication|
    US20030140132A1|2002-01-22|2003-07-24|Yves-Andre Champagne|Method and apparatus for updating network device configuration information in a network management system|
    US20080232274A1|2007-03-22|2008-09-25|Telecommunications Research Laboratories|Distributed synchronous batch reconfiguration of a network|
    WO2002088875A2|2001-04-27|2002-11-07|The Boeing Company|Communicating data through a network|
    US8429630B2|2005-09-15|2013-04-23|Ca, Inc.|Globally distributed utility computing cloud|
    US20130318199A1|2012-04-27|2013-11-28|Hervé LE JOUAN|Managing data on computer and telecommunications networks|
    US20140280913A1|2013-03-15|2014-09-18|Moki Mobility, Inc.|Device and settings management platform|
    KR101794222B1|2013-04-09|2017-11-07|사이트릭스 시스템스, 인크.|Providing a native desktop using cloud-synchronized data|
    US9099162B2|2013-10-31|2015-08-04|Oracle International Corporation|Media and drive validation in tape libraries|
    US10742520B2|2013-12-31|2020-08-11|Citrix Systems, Inc.|Providing mobile device management functionalities|
    US20150334162A1|2014-05-13|2015-11-19|Citrix Systems, Inc.|Navigation of Virtual Desktop Content on Devices|CN105867837A|2015-12-02|2016-08-17|乐视体育文化产业发展（北京）有限公司|Method, equipment and system for updating configurations of clients in distributed high-speed cache systems|
    CN105871584A|2015-12-02|2016-08-17|乐视体育文化产业发展（北京）有限公司|Method, device and system for updating client configuration in key value pair database|
    US11238176B1|2016-06-17|2022-02-01|BigID Inc.|System and methods for privacy management|
    CN108011909B|2016-10-28|2020-09-01|北京市商汤科技开发有限公司|Communication method and system, electronic device and computer cluster|
    法律状态:
    2015-05-26| PLFP| Fee payment|Year of fee payment: 2 |
    2015-12-18| PLSC| Search report ready|Effective date: 20151218 |
    2016-05-26| PLFP| Fee payment|Year of fee payment: 3 |
    2017-05-23| PLFP| Fee payment|Year of fee payment: 4 |
    2018-05-25| PLFP| Fee payment|Year of fee payment: 5 |
    2020-06-26| PLFP| Fee payment|Year of fee payment: 7 |
    2021-06-25| PLFP| Fee payment|Year of fee payment: 8 |
    优先权:
    申请号 | 申请日 | 专利标题
    FR1455446|2014-06-13|
    FR1455446A|FR3022420B1|2014-06-13|2014-06-13|METHODS AND SYSTEMS FOR MANAGING AN INTERCONNECTION NETWORK|FR1455446A| FR3022420B1|2014-06-13|2014-06-13|METHODS AND SYSTEMS FOR MANAGING AN INTERCONNECTION NETWORK|
    EP15161764.4A| EP2955875B1|2014-06-13|2015-03-30|Server, client and system for managing an interconnection network|
    ES15161764T| ES2700929T3|2014-06-13|2015-03-30|Server, client and management system of an interconnection network|
    US14/706,254| US9866437B2|2014-06-13|2015-05-07|Methods and systems of managing an interconnection network|
    JP2015118359A| JP6304499B2|2014-06-13|2015-06-11|Method and system for managing interconnected networks|
    [返回顶部]