WO2017071442A1 - 一种负载分担的方法、装置和*** - Google Patents
一种负载分担的方法、装置和*** Download PDFInfo
- Publication number
- WO2017071442A1 WO2017071442A1 PCT/CN2016/099892 CN2016099892W WO2017071442A1 WO 2017071442 A1 WO2017071442 A1 WO 2017071442A1 CN 2016099892 W CN2016099892 W CN 2016099892W WO 2017071442 A1 WO2017071442 A1 WO 2017071442A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- network device
- target network
- load
- controller
- state
- Prior art date
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000000131 plasma-assisted desorption ionisation Methods 0.000 claims abstract 11
- 230000001934 delay Effects 0.000 claims description 28
- 238000012545 processing Methods 0.000 claims description 19
- 230000000977 initiatory effect Effects 0.000 abstract description 3
- 230000004044 response Effects 0.000 description 56
- 230000000903 blocking effect Effects 0.000 description 21
- 238000010586 diagram Methods 0.000 description 10
- 238000004891 communication Methods 0.000 description 7
- 230000008569 process Effects 0.000 description 5
- 230000005540 biological transmission Effects 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000008878 coupling Effects 0.000 description 2
- 238000010168 coupling process Methods 0.000 description 2
- 238000005859 coupling reaction Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 230000003111 delayed effect Effects 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/12—Avoiding congestion; Recovering from congestion
- H04L47/125—Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1004—Server selection for load balancing
- H04L67/1008—Server selection for load balancing based on parameters of servers, e.g. available memory or workload
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/46—Interconnection of networks
- H04L12/4604—LAN interconnection over a backbone network, e.g. Internet, Frame Relay
- H04L12/462—LAN interconnection over a bridge based backbone
- H04L12/4625—Single bridge functionality, e.g. connection of two networks over a single bridge
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L47/00—Traffic control in data switching networks
- H04L47/10—Flow control; Congestion control
- H04L47/11—Identifying congestion
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L65/00—Network arrangements, protocols or services for supporting real-time applications in data packet communication
- H04L65/40—Support for services or applications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1029—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers using data related to the state of servers by a load balancer
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/104—Peer-to-peer [P2P] networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/2854—Wide area networks, e.g. public data networks
- H04L12/2856—Access arrangements, e.g. Internet access
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L67/00—Network arrangements or protocols for supporting network services or applications
- H04L67/01—Protocols
- H04L67/10—Protocols in which an application is distributed across nodes in the network
- H04L67/1001—Protocols in which an application is distributed across nodes in the network for accessing one among a plurality of replicated servers
- H04L67/1036—Load balancing of requests to servers for services different from user content provisioning, e.g. load balancing across domain name servers
Definitions
- the present invention relates to the field of communications technologies, and in particular, to a load sharing method, apparatus, and system.
- communication between endpoint devices can be implemented by Point-to-Point Protocol (PPP).
- PPP Point-to-Point Protocol
- a PPP session can be established first, and then data communication is performed based on the PPP session.
- multiple user equipments and multiple network devices for accessing user equipment are distributed in the same network.
- the user equipment can select a network device in the network to establish a PPP session with the network device.
- the user equipment and the network device can perform various negotiation and data transmission, so that the user equipment can access the network through the network device.
- the inventors have found through research that in the prior art, a large number of user equipments in a network are concentrated on a part of network devices to implement network access, and another part of network devices are only used to carry access of a small number of user equipments. In the network, the load of some network devices is too large and the load of other network devices is controlled. This situation causes the load of each network device in the network to be unbalanced.
- the technical problem to be solved by the embodiments of the present invention is to provide a method, an apparatus, and a system for load sharing to solve the technical problem that the network equipment in the network is not balanced in the prior art.
- an embodiment of the present invention provides a method for load sharing, where the method includes:
- the controller acquires load parameters of the target network device
- the controller sends the first configuration information to the target network device when the target network device is in a state of being overloaded;
- the first configuration information is used to configure the target network device to be in a first state; the first state indicates that the target network device delays or does not return active discovery when receiving an active discovery PADI message. PADO message.
- the controller acquires load parameters of the target network device, including:
- the controller sends a load parameter request to the target network device
- the controller receives a load parameter returned by the target network device for the load parameter request.
- the load parameter request is an extended multi-part Multipart request message of the OpenFlow Openflow protocol
- the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message includes a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, and the request Body field is empty or carries the device identifier;
- the Mulipart response message includes the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the controller acquires a load parameter of the target target network device, specifically:
- the controller determines, according to the load parameter, whether the target network device is in a state of being overloaded, including:
- the controller searches for a minimum value of load parameters of at least two network devices as a minimum load value; the target network device is included in the at least two network devices;
- the controller calculates a difference between the load parameter and the minimum load value as a load difference value
- the controller determines that the target network device is in a state of being overloaded
- the controller determines the target network when the load difference does not exceed the load threshold
- the device is not in a state where the load is out of limits.
- the method further includes:
- the controller sends second configuration information to the target network device when the target network device is not in a state of being overloaded;
- the second configuration information is used to configure the target network device to be in a second state; the second state is to indicate that the target network device returns a PADO message when receiving a PADI message.
- the method further includes:
- the controller After the controller sends the first configuration information to the target network device, the target network device is recorded as an overload device;
- the target network device After the controller sends the second configuration information to the target network device, the target network device is deleted as a record of the overload device.
- the method further includes:
- the controller determines whether the target network device is a recorded overload device
- the controller sends the first configuration information to the target network device, where the target network device is in a state where the load exceeds the limit and the When the target network device is not the recorded overload device, the controller sends the first configuration information to the target network device;
- the controller sends the second configuration information to the target network device, where the target network device is not in a load overrun state, and the When the target network device is a recorded overload device, the controller sends second configuration information to the target network device.
- the load parameter includes a proportion of users of the target network device, bandwidth utilization, CPU utilization, or remaining CPU processing capability.
- the target network device is a broadband access gateway BNG.
- an embodiment of the present invention provides another method for load sharing, where the method includes:
- the target network device receives the first configuration information sent by the controller
- the target network device configures the first state according to the first configuration information
- the first state indicates that the target network device delays or does not return a PADO packet when receiving the PADI packet
- the first configuration information is that when the controller determines that the target network device is in a load overrun state according to a load parameter of the target network device, the controller sends the controller to the target network.
- the method further includes:
- the target network device When in the first state, the target network device delays or does not return a PADO message to the user equipment.
- the method further includes:
- the target network device returns the load parameter to the controller.
- the load parameter request is an extended Multipart request message of the Openflow protocol
- the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message includes a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, and the request Body field is empty or carries the device identifier;
- the Mulipart response message includes the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the method further includes:
- the target network device actively sends the load parameter to the controller.
- the method further includes:
- the target network device configures the second state according to the second configuration information
- the second state indicates that the target network device returns a PADO message when receiving the PADI message
- the second configuration information is sent by the controller to the target network device when the controller determines that the target network device is not in a load overload state according to a load parameter of the target network device.
- an embodiment of the present invention provides a device for load sharing, where the device includes:
- An obtaining module configured to acquire a load parameter of the target network device
- a first determining module configured to determine, according to the load parameter, whether the target network device is in a state of being overloaded
- a first sending module configured to: when the target network device is in a state where the load is out of limits, Transmitting, by the target network device, first configuration information
- the first configuration information is used to configure the target network device to be in a first state; the first state indicates that the target network device delays or does not return active discovery when receiving an active discovery PADI message. PADO message.
- the obtaining module includes:
- a sending submodule configured to send a load parameter request to the target network device
- a receiving submodule configured to receive a load parameter returned by the target network device for the load parameter request.
- the load parameter request is an extended multi-part Multipart request message of the OpenFlow Openflow protocol
- the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message includes a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, where the request Body field is empty or carries the device identifier;
- the Mulipart response message includes the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the acquiring module is specifically configured to receive a load parameter that is actively sent by the target network device.
- the first determining module includes:
- a calculation submodule configured to calculate a difference between the load parameter and the minimum load value as a load difference value
- a first determining submodule configured to determine whether the load difference exceeds a preset load threshold
- a second determining submodule configured to determine that the target network device is in a state where the load exceeds the limit when the load difference exceeds the load threshold
- a third determining submodule configured to determine that the target network device is not in a state of being overloaded when the load difference does not exceed the load threshold.
- the device further includes:
- a second sending module configured to send second configuration information to the target network device when the target network device is not in a state of being overloaded
- the second configuration information is used to configure the target network device to be in a second state; the second state is to indicate that the target network device returns a PADO message when receiving a PADI message.
- the device further includes:
- a recording module configured to record the target network device as an overload device after sending the first configuration information to the target network device
- the device further includes:
- a second determining module configured to determine whether the target network device is a recorded overload device
- the first sending module is configured to send the first configuration information to the target network device when the target network device is in a load overload state and the target network device is not a recorded overload device;
- the second sending module is configured to send, when the target network device is not in a load overload state, and the target network device is a recorded overload device, the controller sends a second message to the target network device. Configuration information.
- the load parameter includes a proportion of users of the target network device, bandwidth utilization, CPU utilization, or remaining CPU processing capability.
- the target network device is a broadband access gateway BNG.
- an embodiment of the present invention provides another device for load sharing, where the device includes:
- a first receiving module configured to receive first configuration information sent by the controller
- a first configuration module configured to configure a first state according to the first configuration information
- the first state indicates that the target network device delays or does not return a PADO packet when receiving the PADI packet
- the first configuration information is sent by the controller to the target network device when the controller determines that the target network device is in a load overload state according to a load parameter of the target network device.
- the device further includes:
- a second receiving module configured to receive a PADI message sent by the host
- the target network device delays or does not return a PADO message to the host.
- the device further includes:
- a third receiving module configured to receive a load parameter request sent by the controller
- the load parameter request is an extended Multipart request message of the Openflow protocol
- the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message includes a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, and the request Body field is empty or carries the device identifier;
- the Mulipart response message includes the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the device further includes:
- a sending module configured to actively send the load parameter to the controller.
- the device further includes:
- a fourth receiving module configured to receive second configuration information sent by the controller
- a second configuration module configured to configure a second state according to the second configuration information
- the second state indicates that the target network device returns a PADO message when receiving the PADI message
- the second configuration information is sent by the controller to the target network device when the controller determines that the target network device is not in a load overload state according to a load parameter of the target network device.
- the embodiment of the present invention provides a load sharing system, where the system includes a controller and a network device, where the controller is configured with the apparatus of any one of the foregoing third aspects, where the network device is configured. There is a device of any of the foregoing fourth aspects.
- the embodiment of the invention has at least the following advantages:
- the target network device represents any network device in the network that needs load control
- the controller acquires a load parameter of the target network device, and determines, according to the load parameter, whether the target network device is in an overload limit.
- the state if the target network device is in a state where the load is out of limits, the controller sends the first configuration information to the target network device, and the target network device configures the first state when receiving the first configuration information, in the case of being in the first state
- PPPoE Active Discovery Offer PPPoE Active Discovery Offer
- PADI PPPoE Active Discovery Initiation
- Initiation referred to as PADO
- the network device with excessive load is configured to be in the first state.
- the network device with excessive load does not immediately return PADO packets to the user equipment.
- the idle network device can return PADO packets to the user equipment earlier than the overloaded network device. Therefore, when the user equipment requests to establish a PPP session to the network device that returns the PADO packet, the user equipment can be idle with the load.
- the network device establishes a PPP session, so that the user equipment selects the idle network device to access the network, and prevents the user equipment from selecting the network device with excessive load to access the network, so that the load of each network device in the network is more balanced.
- the controller may implement load control on the network device by sending the first configuration information for configuring the first state to the network device, and implementing the network device in the network in this manner.
- Load balancing on the one hand, the load control of the network device can eliminate the need for additional switching devices for controlling the packet forwarding path between the user equipment and the network device, so that load balancing can be implemented without modifying the network architecture, and on the other hand, the network device Load control can eliminate the need to configure the control commands of the packet forwarding path on the user equipment in advance, so that load balancing can be implemented without technical modification of a large number of user equipments. It can be seen that, through the technical solution of the embodiment of the present invention, network load balancing is difficult to implement and low in cost, thereby making network load balancing easier to implement.
- FIG. 1 is a schematic diagram of a network system framework involved in an application scenario according to an embodiment of the present invention
- FIG. 2 is a flowchart of an embodiment of a method for load sharing according to the present invention.
- FIG. 3 is a flowchart of another embodiment of a method for load sharing according to the present invention.
- FIG. 4 is a flowchart of still another embodiment of a method for load sharing according to the present invention.
- FIG. 5 is a flowchart of still another embodiment of a method for load sharing according to the present invention.
- FIG. 6 is a schematic structural diagram of an embodiment of a load sharing device according to the present invention.
- FIG. 7 is a schematic structural view of another embodiment of a load sharing device according to the present invention.
- FIG. 8 is a schematic structural diagram of an embodiment of a load sharing system according to the present invention.
- FIG. 9 is a schematic structural diagram of a controller according to an embodiment of the present invention.
- FIG. 10 is a schematic structural diagram of a target network device according to an embodiment of the present invention.
- the process of user equipment accessing the network for network access includes two phases, namely, a discovery phase and a session phase.
- the discovery phase when a user equipment needs to access the network, the user equipment can send PADI messages to all network devices in the network, and the PADI message is used to request a specific type of service from the network device; each network device is When receiving the PADI message, it can determine whether it can provide the service requested by the PADI message, and if it can provide, the PADO message can be returned to the user equipment; the user equipment often receives the PADI message after receiving the PADI message.
- the user equipment selects to send a PPP session to the network device that returns the PADO packet, and the network device provides a session ID to the user equipment, so that the user equipment A PPP session is established with the network device.
- the user equipment and the network device can implement a PPP session based on the session identifier.
- the user equipment can perform various negotiation and data transmission with the network device, thereby implementing user equipment access to the network. It can be seen that, because the user equipment chooses to establish a PPP session to the network device that returns the PADO packet, some network devices in the network need to carry a large number of user equipments to perform PPP sessions, thereby causing load overload and other networks in the network. The device only needs to carry a PPP session of a small number of user equipments, which causes the load to be idle. Therefore, in the prior art, the load of each network device in the network is unbalanced.
- a load sharing method and device are provided.
- the network device that is overloaded does not immediately return PADO packets to the received PADI packets.
- the network device with idle load can return PADO packets to the user device earlier than the network device with excessive load. Therefore, the user equipment can By loading idle network devices instead of overloaded network devices to access the network, the load of each network device in the network is more balanced.
- one of the application scenarios of the embodiment of the present invention may be applied to a network system composed of a controller 101 and a target network device 102.
- the controller 101 acquires the mesh
- the load parameter of the network device 102 is marked, and according to the load parameter, it is determined whether the target network device 102 is in a state where the load exceeds the limit. Then, if the target network device 102 is in a state where the load is out of limits, the controller 101 sends the first configuration information to the target network device 102. Then, the target network device 102 receives the first configuration information sent by the controller 101, and configures the first state according to the first configuration information.
- the first state indicates that the target network device 102 delays or does not return a PADO message when receiving the PADI message.
- the target network device 102 may be any one of the network systems used to access the user equipment.
- FIG. 2 is a flowchart of an embodiment of a method for load sharing in the present invention.
- the method may specifically include the following steps:
- Step 201 The controller acquires a load parameter of the target network device.
- the controller acquires load parameters of the target network device.
- the target network device may be any network device used to access the user equipment under the control of the controller.
- the target network device may be a switching device corresponding to any port under the control of the controller.
- the target network device may be a switching device corresponding to any slot under the control of the controller.
- the load parameter is a parameter for describing a load status of a target network device.
- the load parameter may be user utilization radio, bandwidth utilization radio, CPU utilization radio, or remaining CPU processing power.
- the user utilization of the target network device indicates a ratio of the number of accessed user devices on the target network device to the total number of user devices.
- the bandwidth utilization of the target network device represents a ratio of the occupied bandwidth on the target network device to the total bandwidth.
- the CPU utilization of the target network device indicates a ratio of CPU processing capacity occupied by the target network device to total CPU processing capability.
- the remaining CPU processing capability of the target network device may be calculated according to the total CPU processing capability and CPU utilization of the target network device.
- the controller can obtain the load parameters in a plurality of different manners. take.
- the controller may actively request load parameters from the target network device, and the target network device returns its own load parameters upon receiving the load parameter request.
- the step 201 includes: the controller sending a load parameter request to the target network device; and the controller receiving a load parameter returned by the target network device to the load parameter request.
- the timing of sending the load parameter to the target network device by the controller is not limited in this embodiment.
- the controller may send a load parameter request to the target network device when the target network device is needed. At this time, the controller performs load control on the target network device according to the load parameter currently returned by the target network device.
- the controller may periodically send a load parameter request to the target network device and save the latest load parameter of the target network device in real time. At this time, the controller loads the target network device according to the load parameter of the target network device saved by the controller. control.
- the target network device can actively send load parameters to the controller.
- step 201 is specifically: the controller receives a load parameter that is actively sent by the target network device.
- the timing of sending the load parameter to the controller by the target network device is not limited in this embodiment.
- the target network device may periodically send the load parameter to the target network device, and the controller saves the latest load parameter of the target network device in real time.
- the controller searches the target network device according to the load parameter of the target network device saved by the controller. Perform load control.
- the transmission of load parameters can be implemented between the target network device and the controller through various protocols, such as OpenFlow protocol, Network Configuration (Netconf) protocol, or Simple Network Management Protocol (Simple). Network Management Protocol (SNMP) protocol.
- OpenFlow OpenFlow protocol
- Netconf Network Configuration
- SNMP Network Management Protocol
- the load parameter request sent by the controller to the target network device may be an extended multipart request message, and the load parameter sent by the target network device to the controller may be sent by using a Multipart response message.
- the extended Multipart request message includes a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, the request Body field is empty or carries the device identifier, and the Muptipart request message is Is a load parameter request.
- the extended Mulipart response message includes the Type field and a response body Body field, the Type field carrying a type value indicating load balancing information, the response Body field carrying a load parameter of the target network device, the Multipart response The message is used to send load parameters.
- the foregoing Multipart request message may be in the following format:
- the above Multipart response message can be in the following format:
- the type value indicating the load balancing information in the type field may be a newly defined type value.
- the type value can be defined as 17, ie:
- the request Body field may be in the following format:
- the port_no field carries the port number of the target network device
- the slot_id field carries the slot identifier of the target network device.
- the response Body field may be in the following format:
- the bandwidth field carries the bandwidth of the port where the target network device is located
- the max_user_num field carries the total number of users in the slot where the target network device is located
- the CPU_capability field carries the CPU of the slot where the target network device is located.
- the mem_size field carries the memory of the slot where the target network device is located
- the bandwidth_uni_ratio field carries the bandwidth utilization of the target network device
- the user_uti_ratio field carries the user utilization of the target network device
- the CPU_uti_ratio carries the CPU utilization of the target network device
- the mem_uti_ratio carries the target network.
- the memory utilization of the device may carry any one or more, that is, the target network device may send any one or more to the controller through a Multipart response message.
- Step 202 The controller determines, according to the load parameter, whether the target network device is in a state where the load exceeds the limit.
- the load state of the target network device may be determined according to the size of the load parameter, and according to the determination result, it may be determined whether the target network device is in a state where the load exceeds the limit.
- the controller can determine and determine the load status of the target network device in a variety of different manners.
- step 202 includes: comparing, by the controller, load parameters of the target network device with a preset load threshold, when comparing The result is that when the load parameter is greater than the load threshold, the controller determines that the target network device is in a state where the load exceeds the limit.
- the controller determines that the target network device is not in a state where the load exceeds the limit.
- the controller generally controls multiple network devices.
- the controller may comprehensively determine whether the target network device is based on a load relationship between the target network device and other network devices. In a state where the load is overrun.
- step 202 includes: the controller searching for a minimum value of load parameters of at least two network devices as a minimum load value; and calculating, by the controller, a difference between the load parameter and the minimum load value And determining, by the controller, whether the load difference exceeds a preset load threshold; when the load difference exceeds the load threshold, the controller determines that the target network device is in a load super a limited state; when the load difference does not exceed the load threshold, the controller determines that the target network device is not in a state of load overrun.
- the at least two network devices include the target network device.
- the load status of the target network device is determined according to the relationship between the load difference and the load threshold, and the load conditions of the target network device and other network devices can be comprehensively considered, so that the user cannot be accessed in the network.
- the device has too many network devices.
- the at least two network devices comprise any two or more than two network devices under the control of the controller.
- the at least two network devices can include all of the network devices under the controller.
- Step 203 When the target network device is in a state where the load exceeds the limit, the controller sends the first configuration information to the target network device.
- the first configuration information is used to configure the target network device to be in a first state; the first state indicates that the target network device delays or does not return active discovery when receiving an active discovery PADI message. PADO message.
- the controller determines that the target network device is in a state of overload of the load according to the load parameter, the controller sends the first configuration information to the target network device, and the target network device configures itself into the first state when receiving the first configuration information.
- the target network device receives the PADI message of the user equipment, the target network device determines whether it is in the first state. If the target network device determines that it is in the first state, the target network device delays returning the PADO message to the user equipment. Or do not return PADO packets to the user device. Therefore, when the target network device is in the overloaded state, the PADO packet received by the user equipment is usually not the PADO packet of the target network device, and the user equipment usually does not choose to establish a PPP session with the target network device. User equipment can Avoid accessing the network through a heavily loaded target network device.
- the embodiment When the controller determines that the target network device is in a state where the load is out of limits, the embodiment provides two embodiments for controlling the load of the target network device.
- the first configuration information is a delay policy that is sent by the controller to the target network device, where the first state indicates a state in which the target network device has configured the delay policy, where the delay policy is used to indicate the target network device to the PADI.
- Message delay processing Specifically, when the target network device receives the delay policy sent by the controller, the delay policy is configured on the target network device, and thereafter, when the target network device receives the PADI message of the user equipment, the target network device follows the delay policy. The PADI message is saved, and the timer delay is processed to process the PADI message. When the delay time set by the timer arrives, the target network device returns the PADO message to the user equipment.
- the user equipment receives the PADO packet returned by the other idle network device before the target network device returns the PADO packet. Therefore, the user equipment does not choose to load too large.
- the target network device establishes a PPP session, but selects other network devices with idle load to establish a PPP session.
- the first configuration information is a blocking policy that is sent by the controller to the target network device, where the first state indicates that the target network device has configured the blocking policy, and the blocking policy is used to indicate the target network device pair.
- PADI packets are discarded, indicating that the target network device does not return PADO packets for PADI packets.
- the target network device receives the blocking policy sent by the controller, the blocking policy is configured on the target network device, and then, when the target network device receives the PADI packet of the user equipment, the target network device follows the blocking policy.
- the PADI packet is discarded, that is, the target network device does not return a PADO packet to the user equipment. It can be seen that when the target network device is in the overloaded state, the user equipment does not receive the PADO packet from the target network device. Therefore, the user equipment does not select the target network device with excessive load to establish a PPP session.
- the target network device is sometimes in a state of overloaded load, and sometimes in a state where the load is idle.
- the load state of the target network device often changes back and forth between the load overrun and the load idle. Therefore, when the target network device changes from the overloaded state to the idle state, the target network device needs to process the PADI packet normally, and no longer delays the PADI packet, and no longer reports the PADI packet. The text is discarded.
- the embodiment may further include: when the target network device is not in a state where the load is out of limits, the controller sends second configuration information to the target network device; and the second configuration information is used to: Configuring the target network device as the first The second state indicates that the target network device immediately returns a PADO message when receiving the PADI message.
- the second configuration information may be a cancellation policy that is sent by the controller to the target network device, and the second state indicates that the target network device has canceled the delay policy and blocking.
- the status of the policy which is used to indicate the configuration of the target network device cancellation delay policy and blocking policy.
- the target network device performs normal processing on the PADI packet, that is, when the target network device receives the PADI packet, the target network device immediately returns to the PADO. Message.
- the controller is usually controlled by a plurality of network devices, and the controllers send different configuration information to different network devices according to different load states of the network devices, and different network devices are in different configuration states.
- the embodiment may further include: after the controller sends the first configuration information to the target network device, the target network device is recorded as an overload device; After the controller sends the second configuration information to the target network device, the target network device is deleted as a record of the overload device. It can be understood that the controller can identify the network device that has been in the first state by recording the overload device.
- the controller may determine that the target network device is currently in the first state, and may determine that the target network device is currently configured with a delay policy or a blocking policy.
- the controller may determine that the target network device is currently in the second state, and may determine that the target network device is currently not configured with a delay policy or a blocking policy.
- the controller determines that the target network device is in a state where the load exceeds the limit, the target network device is sometimes in the first state, and the controller may not send the first configuration information to the target network device.
- the controller determines that the target network device is not in a state where the load is out of limits, the target network device is sometimes in the second state, and the controller does not send the second configuration information to the target network device.
- the embodiment may further include: determining, by the controller, whether the target network device is a recorded overload device; step 203 is specifically: when the target network device is in a state of overloading the load and the target network When the device is not a recorded overload device, the controller sends the first configuration information to the target network device; the foregoing step: “When the target network device is not in a state where the load is out of limits, the controller is The sending, by the target network device, the second configuration information is specifically: when the target network device is not in a state of overloading the load and the target network device is a recorded overload device, the controller is configured to the target network device Send the second configuration information.
- the target network device is recorded as an overloaded device indicating that the target network device is in the first state, and the target network device is not recorded as the overloaded device, indicating that the target network device is in the second state.
- the controller sends the first configuration information to the target network device only when the target network device load of the second state exceeds the limit, and only the target network device in the first state is in the idle state of the load to the target network.
- the device sends the second configuration information. It can be seen that the number of configuration information that needs to be transmitted in the network is small, and the controller and the network device have a lighter processing load on the configuration information.
- the target network device mentioned in this embodiment may be a Broadband Network Gateway (BNG) or an Access Concentrator. More specifically, the target network device may be a Broadband Remote Access Server (BRAS).
- the user equipment mentioned in this embodiment may be a host, or may be a mobile terminal, etc., which is not limited in this embodiment.
- the controller configures the overloaded network device to be in the first state, and when receiving the PADI sent by the user equipment, the configuration of the overloaded network device through the controller is extended.
- the network device with the idle load can return the PADO packet to the user equipment earlier than the overloaded network device. Therefore, when the user equipment requests to establish a PPP session to the network device that returns the PADO packet, the user equipment can establish a PPP session with the network device with the idle load. It can be seen that the user equipment can select the idle network device to access the network, and can avoid the selection. An overloaded network device accesses the network, so the load of each network device in the network is more balanced.
- the controller may implement load control on the network device by sending the first configuration information for configuring the first state to the network device.
- the load balancing of each network device in the network is implemented.
- the load balancing implementation does not need to modify the network architecture.
- the user equipment does not need to pre-configure the control commands of the packet forwarding path. Therefore, the load balancing implementation does not need to modify a large number of user equipments. It can be seen that, through the technical solution of the embodiment of the present invention, network load balancing is difficult to implement and low in cost, and therefore, network load balancing is easier to implement.
- FIG. 3 is a flowchart of another embodiment of a method for load sharing according to the present invention.
- the method may specifically include the following steps:
- Step 301 The target network device receives the first configuration information sent by the controller.
- Step 302 The target network device configures a first state according to the first configuration information.
- the first state indicates that the target network device delays or does not return a PADO packet when receiving the PADI packet
- the first configuration information is sent by the controller to the target network device when the controller determines that the target network device is in a load overload state according to a load parameter of the target network device.
- the embodiment may further include:
- the target network device When in the first state, the target network device delays or does not return a PADO message to the user equipment.
- the embodiment may further include:
- the target network device returns the load parameter to the controller.
- the load parameter request may be an extended Multipart request message of the Openflow protocol
- the load parameter may be sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message may include a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, the request Body field is empty or carries the device identifier;
- the Mulipart response message may include the Type field and a response body Body field, where the response Body field carries a load parameter of the target network device.
- the embodiment may further include:
- the target network device actively sends the load parameter to the controller.
- the embodiment may further include:
- the target network device configures the second state according to the second configuration information
- the second state indicates that the target network device returns a PADO message when receiving the PADI message
- the second configuration information is sent by the controller to the target network device when the controller determines that the target network device is not in a load overload state according to a load parameter of the target network device.
- the content of the load parameter, the first configuration information, the second configuration information, the first state, and the second state in the embodiment of the present invention can be referred to the embodiment shown in FIG. 2 .
- the method for performing the method in the target network device in the embodiment of the present invention reference may be made to the detailed description of the embodiment shown in FIG. 2, and details are not described herein again.
- the network device with excessive load is configured to be in the first state by the controller, and when the PADI sent by the user device is received, the network device with excessive load delays or refuses to the user.
- the device returns PADO packets instead of immediately returning PADO packets to the user equipment. Therefore, the network device with idle load can return PADO packets to the user equipment earlier than the overloaded network device. Therefore, when the user equipment returns to the earliest.
- the network device of the PADO packet requests to establish a PPP session
- the user equipment can establish a PPP session with the network device with the idle load. It can be seen that the user equipment can select the network device with the idle load to access the network, and can avoid selecting the network with excessive load.
- the device accesses the network, so the load of each network device in the network is more balanced.
- the first configuration information is a delay policy, where the first state indicates that the target network device delays the processing of the PADO message to the PADI message, and the second configuration information is the cancellation delay policy.
- the two states indicate that the target network device performs an instant return of the PADO message to the PADI message.
- the controller performs load balancing control on the first network device and the second network device, and the user equipment needs to simultaneously initiate a PADI message to the first network device and the second network device when accessing the network.
- FIG. 4 is a flowchart of still another embodiment of a method for load sharing according to an embodiment of the present invention. This embodiment can be applied to the first exemplary application scenario described above, and the method includes the following steps:
- Step 401 The controller acquires a first load parameter of the first network device and a second load parameter of the second network device.
- the first network device sends the current load parameter of the first network device to the controller as the first load parameter by using the Multipart response message
- the second network device sends the current load of the second network device to the controller by using the Multipart response message.
- the parameter is used as the second load parameter.
- Step 402 The controller determines, according to the first load parameter, that the first network device is in a load overload condition. And determining, according to the record of the overloaded device, that the first network device is not recorded as an overload device, transmitting a delay policy to the first network device, and recording the first network device as an overload device.
- Step 403 The first network device configures the received delay policy.
- Step 404 The controller determines, according to the second load parameter, that the second network device is not in a state of being overloaded, and determines, according to the record of the overload device, that the second network device is not recorded as an overload device, and refuses to send configuration information to the second network device. .
- Step 405 The user equipment sends a PADI message to the first network device and the second network device, respectively.
- Step 406 When the second network device receives the PADI message, it immediately returns a PADO message to the user equipment.
- Step 407 When receiving the PADO packet, the first network device returns the PADO packet to the user equipment after delaying for a period of time according to the configured delay policy.
- Step 408 The user equipment first receives the PADO message returned by the second network device, and then receives the PADO message returned by the first network device, and the user equipment establishes a PPP session with the second network device.
- Step 409 The controller acquires a third load parameter of the first network device and a fourth load parameter of the second network device.
- the first network device sends the current load parameter of the first network device to the controller as the third load parameter again by using the Multipart response message; the second network device sends the second network device to the controller again through the Multipart response message.
- the load parameter is used as the fourth load parameter.
- Step 410 The controller determines, according to the third load parameter, that the first network device is not in a load overrun state, and determines, according to the record of the overload device, that the first network device has been recorded as an overload device, and sends a cancel delay policy to the first network device. And delete the first network device as a record of the overloaded device.
- Step 411 When the first network device receives the cancellation delay policy, delete the configuration of the delay policy.
- Step 412 The controller determines, according to the fourth load parameter, that the second network device is in a state of being overloaded, and determines, according to the record of the overload device, that the second network device is not recorded as an overload device, and sends a delay policy to the second network device, and The second network device is recorded as an overload device.
- Step 413 The second network device configures the received delay policy.
- Step 414 The user equipment sends a PADI message to the first network device and the second network device, respectively.
- Step 415 When the first network device receives the PADI message, it immediately returns a PADO message to the user equipment.
- Step 416 When the second network device receives the PADO packet, according to the configured delay policy, The PADO message is returned to the user equipment after a delay.
- Step 417 The user equipment first receives the PADO message returned by the first network device, and then receives the PADO message returned by the second network device, and the user equipment establishes a PPP session with the first network device.
- the controller sends a delay policy to the network device, and the network device with excessive load delays the processing of the PADO packet by the PADI packet, and the idle network device performs the PADI packet immediately.
- the processing of the PADO packet is returned. Therefore, the user equipment can select to establish a PPP session with the network device with the idle load instead of the overloaded network device, thus implementing load balancing.
- the first configuration information is a blocking policy
- the first state indicates that the target network device performs the discarding process on the PADI packet
- the second configuration information is the unblocking policy
- the second state indicates the target network device.
- the controller performs load balancing control on the first network device and the second network device, and the user equipment needs to simultaneously initiate a PADI message to the first network device and the second network device when accessing the network.
- FIG. 5 is a flowchart of still another embodiment of a method for load sharing according to an embodiment of the present invention. This embodiment can be applied to the second exemplary application scenario described above, and the method includes the following steps:
- Step 501 The controller acquires a first load parameter of the first network device and a second load parameter of the second network device.
- the first network device sends the current load parameter of the first network device to the controller as the first load parameter by using the Multipart response message
- the second network device sends the current load of the second network device to the controller by using the Multipart response message.
- the parameter is used as the second load parameter.
- Step 502 The controller determines, according to the first load parameter, that the first network device is in a state of being overloaded, and determines that the first network device is not recorded as an overload device according to the record of the overload device, and sends a blocking policy to the first network device, and The first network device is recorded as an overload device.
- Step 503 The first network device configures the received blocking policy.
- Step 504 The controller determines, according to the second load parameter, that the second network device is not in the overloaded state, and determines that the second network device is not recorded as the overload device according to the record of the overload device, and refuses to send the configuration information to the second network device. .
- Step 505 The user equipment sends a PADI message to the first network device and the second network device, respectively.
- Step 506 When the second network device receives the PADI message, it immediately returns a PADO message to the user equipment.
- Step 507 When the first network device receives the PADO packet, according to the configured blocking policy, Discard PADI messages.
- the first network device discards the PADI message, that is, the first network device refuses to return the PADO message to the user equipment.
- Step 508 The user equipment receives the PADO packet returned by the second network device, and does not receive the PADO packet returned by the first network device, where the user equipment establishes a PPP session with the second network device.
- Step 509 The controller acquires a third load parameter of the first network device and a fourth load parameter of the second network device.
- the first network device sends the current load parameter of the first network device to the controller as the third load parameter again by using the Multipart response message; the second network device sends the second network device to the controller again through the Multipart response message.
- the load parameter is used as the fourth load parameter.
- Step 510 The controller determines, according to the third load parameter, that the first network device is not in the overloaded state, and determines, according to the record of the overloaded device, that the first network device has been recorded as an overload device, and sends an unblocking policy to the first network device. And delete the first network device as a record of the overloaded device.
- Step 511 When the first network device receives the unblocking policy, the configuration of the blocking policy is deleted.
- Step 512 The controller determines, according to the fourth load parameter, that the second network device is in a state where the load exceeds the limit, and determines, according to the record of the overload device, that the second network device is not recorded as an overload device, and sends a blocking policy to the second network device, and The second network device is recorded as an overload device.
- Step 513 The second network device configures the received blocking policy.
- Step 513 The user equipment sends a PADI message to the first network device and the second network device, respectively.
- Step 514 When the first network device receives the PADI message, it immediately returns a PADO message to the user equipment.
- Step 515 When receiving the PADO packet, the second network device discards the PADI packet according to the configured blocking policy.
- the second network device discards the PADI message, that is, the second network device refuses to return the PADO message to the user equipment.
- Step 515 The user equipment receives the PADO message returned by the first network device and does not receive the PADO message returned by the second network device, and the user equipment establishes a PPP session with the first network device.
- the controller sends a blocking policy to the network device, and the network device with excessive load discards the PADI packet, and the network device with the idle load immediately returns the PADO packet for the PADI packet. , therefore, the user device is able to select a network that is idle with the load The device establishes a PPP session instead of overloading the network device, thus achieving load balancing.
- FIG. 6 is a schematic structural diagram of an embodiment of a load sharing device according to the present invention.
- the device may be configured in the controller 600.
- the device includes:
- the obtaining module 601 is configured to acquire a load parameter of the target network device.
- a first determining module 602 configured to determine, according to the load parameter, whether the target network device is in a state of being overloaded
- the first sending module 603 is configured to: when the target network device is in a state where the load exceeds the limit, send the first configuration information to the target network device;
- the first configuration information is used to configure the target network device to be in a first state; the first state indicates that the target network device delays or does not return active discovery when receiving an active discovery PADI message. PADO message.
- the acquiring module 601 acquires the load parameter of the target network device, and the first determining module 602 determines, according to the load parameter acquired by the obtaining module 601, whether the target network device is in a state where the load exceeds the limit, and the first sending The module 603 sends the first configuration information to the target network device if the first determining module 602 determines that the target network device is in a state where the load is out of limits.
- the obtaining module 601 may include a sending submodule and a receiving submodule.
- the sending submodule is configured to send a load parameter request to the target network device
- the receiving submodule is configured to receive a load parameter returned by the target network device to the load parameter request.
- the load parameter request may be an extended multi-part Multipart request message of the OpenFlow Openflow protocol, and the load parameter may be sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message may include a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, where the request Body field is empty or carries the device identifier; correspondingly,
- the Mulipart response message may include the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the acquiring module 601 may be specifically configured to receive the target network device and send the initiative The load parameter sent.
- the first determining module 602 may include a searching submodule, a computing submodule, a first determining submodule, a second determining submodule, and a third determining submodule.
- the finding submodule is configured to find a minimum value of the load parameters of the at least two network devices as a minimum load value, and a calculation submodule, configured to calculate a difference between the load parameter and the minimum load value, where And a first determining submodule, configured to determine whether the load difference exceeds a preset load threshold; and a second determining submodule, configured to determine, when the load difference exceeds the load threshold, And the third determining submodule is configured to determine that the target network device is not in a state where the load exceeds the limit when the load difference does not exceed the load threshold.
- the at least two network devices include the target network device.
- the device may further include a second sending module.
- the second sending module is configured to: when the target network device is not in a load overload state, send second configuration information to the target network device; the second configuration information is used to The target network device is configured as a second state; the second state indicates that the target network device immediately returns a PADO message when receiving the PADI message.
- the device may further include a recording module and a deleting module.
- the recording module is configured to: after the first configuration information is sent to the target network device, the target network device is recorded as an overload device; and the deleting module is configured to send the second configuration information to the target network device, and then delete The target network device acts as a record of the overloaded device.
- the device may further include a second determining module.
- the second determining module is configured to determine whether the target network device is a recorded overload device, and the first sending module 603 is specifically configured to: when the target network device is in a load overload state and the target When the network device is not the recorded overload device, sending the first configuration information to the target network device; the second sending module is specifically configured to: when the target network device is not in a state of overloading the target and the target When the network device is a recorded overload device, the controller sends the second configuration information to the target network device.
- the load parameter may include a proportion of users of the target network device, bandwidth utilization, CPU utilization, or remaining CPU processing capability.
- the target network device may be a broadband access gateway BNG.
- controller 600 in this embodiment corresponds to the controller mentioned in the foregoing embodiment shown in FIG. 2 .
- the controller 600 delivers the first configuration information to the over-loaded network device, and the overloaded network device is configured to be in the first state.
- the load is If the network device is too large, the configuration of the controller will delay or refuse to return PADO packets to the user equipment instead of immediately returning PADO packets to the user equipment. Therefore, the user equipment can select the network device with idle load to access the network. It can avoid selecting network devices with excessive load to access the network. Therefore, the load of each network device in the network is more balanced.
- FIG. 7 is a schematic structural diagram of another embodiment of a load sharing device according to the present invention.
- the device may be configured in the target network device 700, and the target network device may be any one of the network devices under the control of the controller.
- the device includes:
- the first receiving module 701 is configured to receive first configuration information sent by the controller.
- the first configuration module 702 is configured to configure the first state according to the first configuration information.
- the first state indicates that the target network device delays or does not return a PADO packet when receiving the PADI packet
- the first configuration information is sent by the controller to the target network device when the controller determines that the target network device is in a load overload state according to a load parameter of the target network device.
- the first receiving module 701 receives the first configuration information sent by the controller, and the first configuration module 702 configures the first state according to the first configuration information received by the first receiving module 701. .
- the target network device 700 delays or refuses to return the PADO message when receiving the PADI under the indication of the first state.
- the device may further include a second receiving module and a return control module.
- the second receiving module is configured to receive the PADI message sent by the host
- the returning control module is configured to: when in the first state, the target network device delays or does not return a PADO message to the host.
- the device may further include a third receiving module and a return module.
- the third receiving module is configured to receive a load parameter request sent by the controller, and a return module, configured to return the load parameter to the controller.
- the load parameter request may be an extended Multipart request message of the Openflow protocol, and the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message may include a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, where the request Body field is empty or carries the device identifier; correspondingly,
- the Mulipart response message may include the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the device may further include a sending module.
- the sending module is configured to actively send the load parameter to the controller.
- the device may further include a fourth receiving module and a second configuration module.
- the fourth receiving module is configured to receive the second configuration information that is sent by the controller, and the second configuration module is configured to configure the second state according to the second configuration information, where the second state indicates the target
- the network device receives the PADI message, the network device returns the PADO message; wherein the second configuration information is that the controller determines that the target network device is not in the overload limit according to the load parameter of the target network device.
- the controller sends the status to the target network device.
- the target network device 700 in this embodiment corresponds to the target network device mentioned in the foregoing embodiment shown in FIG. 2 .
- the apparatus configured in the target network device 700 in this embodiment refer to the detailed description of the foregoing embodiment shown in FIG. 2, and details are not described herein again.
- the target network device 700 when the target network device 700 receives the first configuration information sent by the controller when the load is too large, the overloaded network device is configured to be in the first state, and when receiving the PADI sent by the user equipment, When the network device with excessive load is configured by the controller, it will delay or refuse to return PADO packets to the user equipment instead of immediately returning PADO packets to the user equipment. Therefore, the user equipment can select the network device with idle load. The network can avoid selecting a network device with excessive load to access the network. Therefore, the load of each network device in the network is more balanced.
- FIG. 8 is a schematic structural diagram of an embodiment of a load sharing system according to the present invention.
- the system includes a controller 801 and a network device 802.
- the controller 801 is configured with the foregoing apparatus in any one of the embodiments shown in FIG. 6.
- the network device 802 is configured with the foregoing figure.
- the controller 801 corresponds to the controller mentioned in the foregoing embodiment shown in FIG. 2, and the network device 802 in this embodiment corresponds to the one mentioned in the foregoing embodiment shown in FIG. Target network device.
- the controller 801 and the network device 802 in this embodiment For details, refer to the detailed description of the foregoing embodiment shown in FIG. 2, and details are not described herein again.
- the controller configures the overloaded network device to be in the first state.
- the network device with excessive load will delay or reject the configuration of the controller.
- the PADO packet is returned to the user equipment, instead of immediately returning the PADO packet to the user equipment. Therefore, the user equipment can select the network device with the idle load to access the network, and can avoid selecting the network device with excessive load to access the network.
- the load of each network device in the network is more balanced.
- the load balancing of each network device in the network does not require additional switching devices between the user equipment and the network device to control the packet forwarding path. Therefore, there is no need to modify the network architecture or pre-configure packets on the user equipment.
- the control command of the forwarding path does not need to modify a large number of user equipments. It can be seen that the network load balancing is difficult to implement and the cost is low. Therefore, network load balancing is easier to implement.
- FIG. 9 is a schematic structural diagram of a controller according to an embodiment of the present invention.
- the controller 900 includes a processor 901, a memory 902, a network interface 903, and a bus system 904.
- the bus system 904 is for coupling together various hardware components of the controller 900.
- the network interface 903 is configured to implement a communication connection between the controller 900 and at least one other network element, and may use an Internet, a wide area network, a local network, a metropolitan area network, or the like.
- the memory 902 is configured to store program instructions and data.
- the processor 901 is configured to read instructions and data stored in the memory 902, and perform the following operations:
- the processor 901 acquires a load parameter of the target network device
- the processor 901 determines, according to the load parameter, whether the target network device is in a state of being overloaded;
- the processor 901 sends first configuration information to the target network device when the target network device is in a state of being overloaded;
- the first configuration information is used to configure the target network device to be in a first state; the first state indicates that the target network device delays or does not return active discovery when receiving an active discovery PADI message. PADO message.
- the processor 901 may perform the following operations:
- the processor 901 sends a load parameter request to the target network device
- the processor 901 receives a load parameter returned by the target network device for the load parameter request.
- the load parameter request may be an extended multi-part Multipart request message of the OpenFlow Openflow protocol, and the load parameter may be sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message may include a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, where the request Body field is empty or carries the device identifier; correspondingly,
- the Mulipart response message may include the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- the processor 901 may perform the following operations:
- the processor 901 receives a load parameter that is actively sent by the target network device.
- the processor 901 may perform the following operations, in order to determine whether the target network device is in a state of being overloaded according to the load parameter:
- the processor 901 searches for a minimum value of load parameters of at least two network devices as a minimum load value; the target network device is included in the at least two network devices;
- the processor 901 calculates a difference between the load parameter and the minimum load value as a load difference value
- the processor 901 determines whether the load difference exceeds a preset load threshold
- the processor 901 determines that the target network device is in a state of being overloaded
- the processor 901 determines that the target network device is not in a state where the load is out of limits.
- processor 901 can also perform the following operations:
- the processor 901 sends second configuration information to the target network device when the target network device is not in a overloaded state;
- the second configuration information is used to configure the target network device to be in a second state; the second state is to indicate that the target network device returns a PADO message when receiving a PADI message.
- processor 901 can also perform the following operations:
- the processor 901 sends the first configuration information to the target network device, the target network device is recorded as an overload device;
- the target network device is deleted as a record of the overload device.
- processor 901 can also perform the following operations:
- the processor 901 determines whether the target network device is a recorded overload device
- the processor 901 is configured to: when the target network device is in a load overload state, send the first configuration information to the target network device, where the target network device is in a load overload state and When the target network device is not a recorded overload device, the processor 901 sends first configuration information to the target network device;
- the processor sends the second configuration information to the target network device when the target network device is not in the overloaded state, and specifically performs the following operations: when the target network device is not in a load overload state And when the target network device is a recorded overload device, the processor 901 sends the second configuration information to the target network device.
- the load parameter may include a proportion of users of the target network device, bandwidth utilization, CPU utilization, or remaining CPU processing capability.
- the target network device may be a broadband access gateway BNG.
- controller 900 in this embodiment corresponds to the controller in the foregoing embodiment shown in FIG. 2 .
- controller 900 in this embodiment refers to the detailed description of the foregoing embodiment shown in FIG. 2, and details are not described herein again.
- FIG. 10 is a schematic structural diagram of a target network device according to an embodiment of the present invention.
- the target network device 1000 includes a processor 1001, a memory 1002, a network interface 1003, and a bus system 1004.
- the bus system 1004 is for coupling together various hardware components of the controller 1000.
- the network interface 1003 is configured to implement a communication connection between the controller 1000 and at least one other network element, and may use an Internet, a wide area network, a local network, a metropolitan area network, or the like.
- the memory 1002 is configured to store program instructions and data.
- the processor 1001 is configured to read instructions and data stored in the memory 1002, and perform the following operations:
- the processor 1001 receives first configuration information sent by a controller
- the processor 1001 configures a first state according to the first configuration information
- the first state indicates that the target network device delays or does not return a PADO packet when receiving the PADI packet
- the first configuration information is sent by the controller to the target network device when the controller determines that the target network device is in a load overload state according to a load parameter of the target network device.
- processor 1001 may also perform the following operations:
- the processor 1001 receives a PADI message sent by the user equipment
- the processor 1001 delays or does not return a PADO message to the user equipment.
- processor 1001 may also perform the following operations:
- the processor 1001 receives a load parameter request sent by the controller
- the processor 1001 returns the load parameter to the controller.
- the load parameter request may be an extended Multipart request message of the Openflow protocol, and the load parameter is sent by using an extended Multipart response message of the Openflow protocol.
- the Mulicpart request message may include a type Type field and a request body Body field, where the Type field carries a type value indicating load balancing information, where the request Body field is empty or carries the device identifier; correspondingly,
- the Mulipart response message may include the Type field and a response body Body field, and the response Body field carries a load parameter of the target network device.
- processor 1001 may also perform the following operations:
- the processor 1001 actively sends the load parameter to the controller.
- processor 1001 may also perform the following operations:
- the processor 1001 receives second configuration information sent by the controller
- the processor 1001 configures a second state according to the second configuration information
- the second state indicates that the target network device returns a PADO message when receiving the PADI message
- the second configuration information is sent by the controller to the target network device when the controller determines that the target network device is not in a load overload state according to a load parameter of the target network device.
- the target network device 1000 corresponds to the foregoing shown in FIG. 2 Target network device in the example.
- the specific implementations of the target network device 1000 in this embodiment refer to the detailed description of the foregoing embodiment shown in FIG. 2, and details are not described herein again.
- the processor in the embodiment of the present invention may be an integrated circuit chip with signal processing capability.
- each step of the above method may be completed by an integrated logic circuit of hardware in a processor or an instruction in a form of software.
- These instructions may be implemented and controlled by a processor therein for performing the methods disclosed in the embodiments of the present invention.
- the processor may also be a general purpose processor, a digital signal processing (DSP), an application specific integrated circuit, a Field Programmable Gate Array (FPGA) or other programmable logic. Devices, discrete gates or transistor logic devices, discrete hardware components.
- DSP digital signal processing
- FPGA Field Programmable Gate Array
- the above general-purpose processor may be a microprocessor or the processor may be any conventional processor, decoder or the like.
- the steps of the method disclosed in the embodiments of the present invention may be directly implemented as a hardware processor, or may be performed by a combination of hardware and software modules in the processor.
- the software module can be located in a conventional storage medium such as random access memory, flash memory, read only memory, programmable read only memory or electrically erasable programmable memory, registers, and the like.
- bus system may include a power bus, a control bus, and a status signal bus in addition to the data bus.
- the various buses are labeled as a bus system in Figures 9 and 10.
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Computer Hardware Design (AREA)
- General Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
本发明实施例公开了一种负载分担的方法、装置和***。该方法包括:控制器获取目标网络设备的负载参数;所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息;所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。通过本发明实施例,不仅能够使得网络中各网络设备的负载更加均衡,而且网络负载均衡的实现难度小、成本低。
Description
本申请要求于2015年10月28日提交中国专利局、申请号为CN 201510712755.9、发明名称为“一种负载分担的方法、装置和***”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。
本发明涉及通信技术领域,特别是涉及一种负载分担的方法、装置和***。
在网络中,端点设备之间的通信可以通过点到点协议(Point-to-Point Protocol,简称PPP)来实现。两个端点设备之间需要通信时,可以先建立PPP会话,然后再基于PPP会话进行数据通信。在网络接入技术中,同一网络内分布有多个用户设备和多个用于接入用户设备的网络设备。当一个用户设备需要接入网络时,该用户设备可以选择网络中的一个网络设备,与该网络设备建立PPP会话。基于该PPP会话,该用户设备与该网络设备之间可以进行各项协商及数据传输,从而使得该用户设备可以通过该网络设备对网络进行访问。
发明人经过研究发现,现有技术中,网络中的大量的用户设备会集中到一部分网络设备上实现网络接入,而另一部分网络设备却仅用于承载少量用户设备的接入,因此,在网络中,往往一部分网络设备的负载过大而另一部分网络设备的负载控制,这种情况造成了网络中各网络设备的负载不均衡。
发明内容
本发明实施例所要解决的技术问题是,提供一种负载分担的方法、装置和***,以解决现有技术中网络中网络设备负载不均衡的技术问题。
第一方面,本发明实施例提供了一种负载分担的方法,该方法包括:
控制器获取目标网络设备的负载参数;
所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超
限的状态;
当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息;
所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
可选的,所述控制器获取目标网络设备的负载参数,包括:
所述控制器向所述目标网络设备发送负载参数请求;
所述控制器接收所述目标网络设备对所述负载参数请求而返回的负载参数。
可选的,所述负载参数请求是开放流Openflow协议的扩展的多部分Multipart请求消息,所述负载参数通过Openflow协议的扩展的Multipart应答消息进行发送。
可选的,
所述Mulipart请求消息包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;
相应的,所述Mulipart应答消息包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述控制器获取目标目标网络设备的负载参数,具体为:
接收所述目标网络设备主动发送的负载参数。
可选的,所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态,包括:
所述控制器查找至少两个网络设备的负载参数中的最小值,作为最小负载值;所述至少两个网络设备中包括所述目标网络设备;
所述控制器计算所述负载参数与所述最小负载值之间的差值,作为负载差值;
所述控制器确定所述负载差值是否超过预设的负载阈值;
当所述负载差值超过所述负载阈值时,所述控制器确定所述目标网络设备处于负载超限的状态;
当所述负载差值未超过所述负载阈值时,所述控制器确定所述目标网络
设备未处于负载超限的状态。
可选的,所述方法还包括:
当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标网络设备发送第二配置信息;
所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
可选的,所述方法还包括:
所述控制器向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;
所述控制器向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
可选的,所述方法还包括:
所述控制器确定所述目标网络设备是否为已记录的过载设备;
所述当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息,具体为:当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,所述控制器向所述目标网络设备发送第一配置信息;
所述当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标网络设备发送第二配置信息,具体为:当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置信息。
可选的,所述负载参数包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
可选的,所述目标网络设备为宽带接入网关BNG。
第二方面,本发明实施例提供了另一种负载分担的方法,该方法包括:
目标网络设备接收控制器发送的第一配置信息;
所述目标网络设备根据所述第一配置信息,配置第一状态;
所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;
其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网
络设备发送的。
可选的,所述方法还包括:
所述目标网络设备接收用户设备发送的PADI报文;
当处于第一状态时,所述目标网络设备延时或者不向所述用户设备返回PADO报文。
可选的,所述方法还包括:
所述目标网络设备接收所述控制器发送的负载参数请求;
所述目标网络设备向所述控制器返回所述负载参数。
可选的,所述负载参数请求是Openflow协议的扩展的Multipart请求消息,所述负载参数通过Openflow协议的的扩展的Multipart应答消息进行发送。
可选的,
所述Mulipart请求消息包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;
相应的,所述Mulipart应答消息包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述方法还包括:
所述目标网络设备主动向所述控制器发送所述负载参数。
可选的,所述方法还包括:
所述目标网络设备接收所述控制器发送的第二配置信息;
所述目标网络设备根据所述第二配置信息,配置第二状态;
所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;
其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
第三方面,本发明实施例提供了一种负载分担的装置,所述装置包括:
获取模块,用于获取目标网络设备的负载参数;
第一确定模块,用于根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;
第一发送模块,用于当所述目标网络设备处于负载超限的状态时,向所
述目标网络设备发送第一配置信息;
所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
可选的,所述获取模块包括:
发送子模块,用于向所述目标网络设备发送负载参数请求;
接收子模块,用于接收所述目标网络设备对所述负载参数请求而返回的负载参数。
可选的,所述负载参数请求是开放流Openflow协议的扩展的多部分Multipart请求消息,所述负载参数通过Openflow协议的扩展的Multipart应答消息进行发送。
可选的,所述Mulipart请求消息包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;
相应的,所述Mulipart应答消息包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述获取模块,具体用于接收所述目标网络设备主动发送的负载参数。
可选的,所述第一确定模块包括:
查找子模块,用于查找至少两个网络设备的负载参数中的最小值,作为最小负载值;所述至少两个网络设备中包括所述目标网络设备;
计算子模块,用于计算所述负载参数与所述最小负载值之间的差值,作为负载差值;
第一确定子模块,用于确定所述负载差值是否超过预设的负载阈值;
第二确定子模块,用于当所述负载差值超过所述负载阈值时,确定所述目标网络设备处于负载超限的状态
第三确定子模块,用于当所述负载差值未超过所述负载阈值时,确定所述目标网络设备未处于负载超限的状态。
可选的,所述装置还包括:
第二发送模块,用于当所述所述目标网络设备未处于负载超限的状态时,向所述目标网络设备发送第二配置信息;
所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
可选的,所述装置还包括:
记录模块,用于向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;
删除模块,用于向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
可选的,所述装置还包括:
第二确定模块,用于确定所述目标网络设备是否为已记录的过载设备;
所述第一发送模块,具体用于当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,向所述目标网络设备发送第一配置信息;
所述第二发送模块,具体用于当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置信息。
可选的,所述负载参数包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
可选的,所述目标网络设备为宽带接入网关BNG。
第四方面,本发明实施例提供了另一种负载分担的装置,所述装置包括:
第一接收模块,用于接收控制器发送的第一配置信息;
第一配置模块,用于根据所述第一配置信息,配置第一状态;
所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;
其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
可选的,所述装置还包括:
第二接收模块,用于接收主机发送的PADI报文;
返回控制模块,用于当处于第一状态时,所述目标网络设备延时或者不向所述主机返回PADO报文。
可选的,所述装置还包括:
第三接收模块,用于接收所述控制器发送的负载参数请求;
返回模块,用于向所述控制器返回所述负载参数。
可选的,所述负载参数请求是Openflow协议的扩展的Multipart请求消息,所述负载参数通过Openflow协议的的扩展的Multipart应答消息进行发送。
可选的,
所述Mulipart请求消息包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;
相应的,所述Mulipart应答消息包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述装置还包括:
发送模块,用于主动向所述控制器发送所述负载参数。
可选的,所述装置还包括:
第四接收模块,用于接收所述控制器发送的第二配置信息;
第二配置模块,用于根据所述第二配置信息,配置第二状态;
所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;
其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
第五方面,本发明实施例提供了一种负载分担的***,所述***包括控制器和网络设备,所述控制器配置有前述第三方面任意一种实施方式的装置,所述网络设备配置有前述第四方面任意一种实施方式的装置。
与现有技术相比,本发明实施例至少具有以下优点:
采用本发明实施例的技术方案,以目标网络设备表示网络内任意一个需要负载控制的网络设备,控制器获取目标网络设备的负载参数,并根据该负载参数,确定目标网络设备是否处于负载超限的状态,若目标网络设备处于负载超限的状态,控制器向目标网络设备发送第一配置信息,目标网络设备当接收到第一配置信息时配置第一状态,在处于第一状态的情况下,目标网络设备接收到的主动发现发起(PPPoE Active Discovery Initiation,简称PADI)报文时,延时或者不返回主动发现提供(PPPoE Active Discovery Offer
Initiation,简称PADO)报文。由此可见,在网络中,负载过大的网络设备会被配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备不会立即向用户设备返回PADO报文,负载空闲的网络设备则可以早于负载过大的网络设备向用户设备返回PADO报文,因此,当用户设备向最早返回PADO报文的网络设备请求建立PPP会话时,用户设备就可以与负载空闲的网络设备建立PPP会话,从而实现了用户设备选择负载空闲的网络设备接入网络,避免了用户设备选择负载过大的网络设备接入网络,使得网络中各网络设备的负载更加均衡。
此外,本发明实施例的技术方案,控制器可以通过向网络设备下发用于配置第一状态的第一配置信息来实现对网络设备的负载控制,通过这种方式实现网络中各网络设备的负载均衡,一方面网络设备的负载控制可以无需借助用户设备与网络设备之间额外增加用于控制报文转发路径的交换设备,因而无需改造网络架构就可以实现负载均衡,另一方面网络设备的负载控制可以无需预先在用户设备上配置报文转发路径的控制命令,因而无需对大量的用户设备进行技术改造就可以实现负载均衡。由此可见,通过本发明实施例的技术方案,网络负载均衡的实现难度小、成本低,从而使得网络负载均衡更容易实现。
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍,显而易见地,下面描述中的附图仅仅是本申请中记载的一些实施例,对于本领域普通技术人员来讲,还可以根据这些附图获得其他的附图。
图1为本发明实施例中一应用场景所涉及的网络***框架示意图;
图2为本发明中负载分担的方法一实施例的流程图;
图3为本发明中负载分担的方法另一实施例的流程图;
图4为本发明中负载分担的方法又一实施例的流程图;
图5为本发明中负载分担的方法又一实施例的流程图;
图6为本发明中负载分担的装置一实施例的结构示意图;
图7为本发明中负载分担的装置另一实施例的结构示意图;
图8为本发明中负载分担的***一实施例的结构示意图;
图9为本发明实施例中一种控制器的结构示意图;
图10为本发明实施例中一种目标网络设备的结构示意图。
为了使本技术领域的人员更好地理解本发明中的方案,下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清楚、完整地描述,显然,所描述的实施例仅是本发明一部分实施例,而不是全部的实施例。
发明人经过研究发现,现有技术中,用户设备接入网络进行网络访问的过程包括两个阶段,即发现阶段和会话阶段。在发现阶段:当一个用户设备需要接入网络时,该用户设备可以向网络中所有的网络设备发送PADI报文,该PADI报文用于向网络设备请求特定类型的服务;每一个网络设备在接收到该PADI报文时,可以判断自身是否能够提供该PADI报文请求的服务,如果能够提供,则可以向该用户设备返回PADO报文;该用户设备在发送PADI报文之后,往往会收到多个网络设备返回的PADO报文,该用户设备选择向最早返回PADO报文的网络设备请求建立PPP会话,该网络设备再向该用户设备提供会话标识(Session ID),这样,该用户设备与该网络设备之间就建立了PPP会话。在会话阶段:用户设备与网络设备可以基于会话标识实现PPP会话;基于PPP会话,用户设备可以与网络设备进行各项协商及数据传输,从而实现用户设备对网络的访问。由此可见,由于用户设备选择向最早返回PADO报文的网络设备请求建立PPP会话,网络中的一些网络设备需要承载大量用户设备进行PPP会话,从而造成负载超限,而网络中的另一些网络设备则仅需要承载少量用户设备的PPP会话,从而导致负载空闲,因此,现有技术中,网络中各网络设备的负载是不均衡的。
在本发明实施例中,为了使得网络中各网络设备的负载更加均衡,提供了一种负载分担的方法和设备。其中,负载过大的网络设备不会立即对接收到的PADI报文返回PADO报文,负载空闲的网络设备可以早于负载过大的网络设备向用户设备返回PADO报文,因此,用户设备可以通过负载空闲的网络设备而不是负载过大的网络设备来接入网络,网络中各网络设备的负载得以更加均衡。
参见图1,本发明实施例的应用场景之一,可以是应用到由控制器101与目标网络设备102组成的网络***中。在该网络***中,控制器101获取目
标网络设备102的负载参数,并根据所述负载参数,确定所述目标网络设备102是否处于负载超限的状态。然后,若所述目标网络设备102处于负载超限的状态,所述控制器101向所述目标网络设备102发送第一配置信息。再后,所述目标网络设备102接收所述控制器101发送的第一配置信息,并根据所述第一配置信息配置第一状态。所述第一状态指示所述目标网络设备102接收到PADI报文时,延时或不返回PADO报文。其中,目标网络设备102可以是网络***中的任意一个用于接入用户设备的网络设备。
需要注意的是,上述应用场景仅是为了便于理解本发明实施例的原理而示出,本发明实施例在此方面不受任何限制。相反,本发明实施例可以应用于适用的任何场景。
下面结合附图,通过具体的实施例来详细说明本发明中负载分担的方法、装置和***的实现方式。
参见图2,图2是本发明中负载分担的方法一实施例的流程图。在本实施例中,所述方法例如具体可以包括如下步骤:
步骤201、控制器获取目标网络设备的负载参数。
当控制器需要对目标网络设备进行负载控制时,控制器获取目标网络设备的负载参数。其中,所述目标网络设备可以是在所述控制器控制下的任意一个用于接入用户设备的网络设备。例如,目标网络设备可以是在控制器控制下任意一个端口(port)对应的交换设备。又如,目标网络设备可以是在控制器控制下任意一个槽位(slot)对应的交换设备。所述负载参数是用于描述目标网络设备的负载状态的参数。例如,该负载参数可以是用户利用率(user utilization radio)、带宽利用率(bandwidth utilization radio)、CPU利用率(CPU utilization radio)或剩余CPU处理能力。所述目标网络设备的用户利用率表示所述目标网络设备上已接入用户设备的数量占总用户设备的数量的比例。所述目标网络设备的带宽利用率表示所述目标网络设备上已占用的带宽占总带宽的比例。所述目标网络设备的CPU利用率表示所述目标网络设备上已占用的CPU处理能力占总CPU处理能力的比例。所述目标网络设备的剩余CPU处理能力可以根据所述目标网络设备的总CPU处理能力与CPU利用率进行计算。
在具体实施方式中,控制器可以采用多种不同的方式对负载参数进行获
取。
作为一种示例,控制器可以主动向目标网络设备请求负载参数,目标网络设备则在接收到负载参数请求时返回自身的负载参数。具体地,步骤201包括:所述控制器向所述目标网络设备发送负载参数请求;所述控制器接收所述目标网络设备对所述负载参数请求而返回的负载参数。其中,对控制器向目标网络设备发送负载参数的时机,本实施例不作限定。例如,控制器可以在需要对目标网络设备时向目标网络设备发送负载参数请求,此时,控制器根据目标网络设备当前返回的负载参数对目标网络设备进行负载控制。或者,控制器也可以定期地向目标网络设备发送负载参数请求并实时地保存目标网络设备最新返回的负载参数,此时,控制器根据自身保存的目标网络设备的负载参数对目标网络设备进行负载控制。
作为另一种示例,目标网络设备可以主动向控制器发送负载参数。此时,步骤201具体为:所述控制器接收所述目标网络设备主动发送的负载参数。其中,对于目标网络设备向控制器发送负载参数的时机,本实施例不作限定。例如,目标网络设备可以定期地主动向目标网络设备发送负载参数,控制器则实时地保存目标网络设备最新返回的负载参数,此时,控制器根据自身保存的目标网络设备的负载参数对目标网络设备进行负载控制。
在上述两种示例中,目标网络设备与控制器之间可以通过多种不同的协议来实现负载参数的传输,如开放流(Openflow)协议、网络配置(Netconf)协议或简单网络管理协议(Simple Network Management Protocol,简称SNMP)协议。
作为一种示例,在Openflow协议中,控制器向目标网络设备发送的负载参数请求可以是扩展的多部分(Multipart)请求消息,目标网络设备向控制器发送的负载参数可以通过Multipart应答消息发送。具体地,扩展的Multipart请求消息包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识,此Muptipart请求消息即是一个负载参数请求。相应的,扩展的Mulipart应答消息包括所述Type字段和应答主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述应答Body字段携带所述目标网络设备的负载参数,此Multipart应答消息即用于发送负载参数。
在具体实施方式中,上述Multipart请求消息可以采用如下格式:
上述Multipart应答消息可以采用如下格式:
在上述Multipart请求消息和Multipart应答消息中,位于type字段的指示负载均衡信息的类型值,可以是一个新定义的类型值。例如,该类型值可以定义为17,即:
OFPMP_BRAS_LOADBALANCE_INFO=17
在上述Multipart请求消息中,请求Body字段可以采用如下格式:
在发送给目标网络设备的Multipart请求消息的Body字段中,port_no字段携带目标网络设备的端口号,slot_id字段携带目标网络设备的槽位标识。
在上述Multipart应答消息中,应答Body字段可以采用如下格式:
在目标网络设备发送的Multipart应答消息的Body字段中,bandwidth字段携带目标网络设备所在端口的带宽,max_user_num字段携带目标网络设备所在槽位的用户总数,CPU_capability字段携带目标网络设备所在槽位的CPU处理能力,mem_size字段携带目标网络设备所在槽位的内存,bandwidth_uni_ratio字段携带目标网络设备的带宽利用率,user_uti_ratio字段携带目标网络设备的用户利用率,CPU_uti_ratio携带目标网络设备的CPU利用率,mem_uti_ratio携带目标网络设备的内存利用率。可以理解的是,在上述这些字段中,目标网络设备向控制器发送的Multipart应答消息可以携带任意一个或多个,即,目标网络设备通过一个Multipart应答消息可以向控制器发送任意一种或多种负载参数。
步骤202、所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态。
当控制器获取到目标网络设备的负载参数时,可以根据负载参数的大小判断目标网络设备的负载状态,根据判断结果可以确定目标网络设备是否处于负载超限的状态。
在具体实施方式中,控制器可以采用多种不同的方式对目标网络设备的负载状态进行判断和确定。
作为一种示例,控制器可以将负载参数与一个预设的阈值相比较,根据比较结果来确定目标网络设备是否处于负载超限的状态。具体地,步骤202包括:控制器将目标网络设备的负载参数与预设的负载阈值相比较,当比较
结果为负载参数大于负载阈值时,控制器确定目标网络设备处于负载超限的状态,当比较结果为负载参数小于负载阈值时,控制器确定目标网络设备未处于负载超限的状态。
作为另一种示例,控制器通常控制有多个网络设备,当需要判断目标网络设备的负载状态时,控制器可以综合根据目标网络设备与其他网络设备之间的负载关系,确定目标网络设备是否处于负载超限的状态。具体地,步骤202包括:所述控制器查找至少两个网络设备的负载参数中的最小值,作为最小负载值;所述控制器计算所述负载参数与所述最小负载值之间的差值,作为负载差值;所述控制器确定所述负载差值是否超过预设的负载阈值;当所述负载差值超过所述负载阈值时,所述控制器确定所述目标网络设备处于负载超限的状态;当所述负载差值未超过所述负载阈值时,所述控制器确定所述目标网络设备未处于负载超限的状态。其中,所述至少两个网络设备包括有所述目标网络设备。可以理解的是,根据负载差值与负载阈值之间的大小关系来确定目标网络设备的负载状态,目标网络设备与其他网络设备的负载情况能够被综合考虑,这样可以避免网络内无法接入用户设备的网络设备数量过多。此外,所述至少两个网络设备包括处于所述控制器控制下的任意两个或任意两个以上的网络设备。例如,所述至少两个网络设备可以包括所述控制器下的所有网络设备。
步骤203、当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息。
所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
当控制器根据负载参数确定目标网络设备处于负载超限的状态时,控制器向目标网络设备发送第一配置信息,目标网络设备在接收到第一配置信息时将自身配置成第一状态。当目标网络设备接收到某用户设备的PADI报文时,目标网络设备确定自身是否处于第一状态,如果目标网络设备确定自身处于第一状态,目标网络设备延时向该用户设备返回PADO报文或者不向该用户设备返回PADO报文。因此,在目标网络设备处于负载超限的状态时,用户设备最早接收到的PADO报文通常不会是目标网络设备的PADO报文,用户设备通常不会选择与目标网络设备建立PPP会话,因此,用户设备可以
避免通过负载过大的目标网络设备接入网络。
当控制器确定目标网络设备处于负载超限的状态时,本实施例提供了两种实施方式对目标网络设备的负载进行控制。
在一种实施方式中,第一配置信息是控制器向目标网络设备下发的延迟策略,第一状态表示目标网络设备已配置该延迟策略的状态,该延迟策略用于指示目标网络设备对PADI报文延时处理。具体地,当目标网络设备接收到控制器下发的延迟策略时,在目标网络设备上配置该延迟策略,此后,当目标网络设备接收到用户设备的PADI报文时,目标网络设备按照延迟策略,保存该PADI报文,并启动定时器延时处理PADI报文,当定时器设置的延迟时间到达时,目标网络设备再向用户设备返回PADO报文。可见,当目标网络设备处于负载超限的状态时,用户设备在目标网络设备返回PADO报文之前会接收到其他负载空闲的网络设备返回的PADO报文,因此,用户设备不会选择负载过大的目标网络设备建立PPP会话,而是会选择其他负载空闲的网络设备建立PPP会话。
在另一种实施方式中,第一配置信息是控制器向目标网络设备下发的阻塞策略,第一状态表示目标网络设备已配置该阻塞策略的状态,该阻塞策略用于指示目标网络设备对PADI报文做丢弃处理,即指示目标网络设备不对PADI报文返回PADO报文。具体地,当目标网络设备接收到控制器下发的阻塞策略时,在目标网络设备上配置该阻塞策略,此后,当目标网络设备接收到用户设备的PADI报文时,目标网络设备按照阻塞策略,对该PADI报文做丢弃处理,也即,目标网络设备不向用户设备返回PADO报文。可见,当目标网络设备处于负载超限的状态时,用户设备不会接收到目标网络设备返回PADO报文,因此,用户设备不会选择负载过大的目标网络设备建立PPP会话。
在具体实施方式中,目标网络设备有时处于负载超限的状态,有时又处于负载空闲的状态,目标网络设备的负载状态常常在负载超限与负载空闲之间来回变化。因此,当目标网络设备从负载超限的状态变成负载空闲的状态时,目标网络设备需要对PADI报文进行正常处理,而不再对PADI报文做延时处理,也不再对PADI报文做丢弃处理。为此,本实施例还可以包括:当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标网络设备发送第二配置信息;所述第二配置信息,用于将所述目标网络设备配置为第
二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。具体地,对于已配置延迟策略或已配置阻塞策略的目标网络网络设备,第二配置信息可以是控制器向目标网络设备下发的取消策略,第二状态表示目标网络设备已取消延迟策略和阻塞策略的状态,该取消策略用于指示目标网络设备取消延迟策略和阻塞策略的的配置。可以理解的是,在目标网络设备取消延迟策略和阻塞策略的情况下,目标网络设备对PADI报文进行正常处理,也即,当目标网络设备接收到PADI报文时,目标网络设备即时返回PADO报文。
在具体实施方式中,控制器通常控制有多个网络设备,而根据网络设备负载状态的不同,控制器向不同的网络设备下发了不同的配置信息,不同的网络设备处于不同的配置状态。为了使得控制器能够识别各网络设备当前处于的配置状态,本实施例还可以包括:所述控制器向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;所述控制器向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。可以理解的是,控制器通过过载设备的记录,可以识别到已处于第一状态的网络设备。当控制器识别到目标网络设备当前被记录为过载设备,控制器可以确定目标网络设备当前已处于第一状态,可以确定目标网络设备当前已配置延迟策略或阻塞策略。当控制器识别到目标网络设备当前未被记录为过载设备,控制器可以确定目标网络设备当前处于第二状态,可以确定目标网络设备当前既未配置延迟策略也未配置阻塞策略。
在具体实施方式中,当控制器确定目标网络设备处于负载超限的状态时,目标网络设备有时已处于为第一状态,此时控制器可以不向目标网络设备发送第一配置信息。类似的,当控制器确定目标网络设备未处于负载超限的状态时,目标网络设备有时已处于第二状态,此时控制器不向目标网络设备发送第二配置信息。具体地,本实施例还可以包括:所述控制器确定所述目标网络设备是否为已记录的过载设备;步骤203具体为:当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,所述控制器向所述目标网络设备发送第一配置信息;前述步骤“当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标网络设备发送第二配置信息”具体为:当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置
信息。其中,目标网络设备被记录为过载设备表明目标网络设备已处于第一状态,目标网络设备未被记录为过载设备表明目标网络设备已处于第二状态。可以理解的是,控制器仅在第二状态的目标网络设备负载超限的情况下向目标网络设备发送第一配置信息,仅在第一状态的目标网络设备处于负载空闲的情况下向目标网络设备发送第二配置信息,可见,网络中需要传输的配置信息数量较少,控制器及网络设备对配置信息的处理负担较轻。
需要说明的是,本实施例提及的目标网络设备,可以是宽带接入网关(Broadband Network Gateway,简称BNG),或者,接入集中器(Access Concentrator)。更具体地,目标网络设备可以是宽带远程接入服务器(Broadband Remote Access Server,简称BRAS)。此外,本实施例提及的用户设备,可以是主机(host),或者也可以是移动终端等,本实施例对此不作限定。
通过本实施例的技术方案,在网络中,控制器将负载过大的网络设备配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备经过控制器的配置将延时或拒绝向用户设备返回PADO报文,而不是立即向用户设备返回PADO报文,因此,负载空闲的网络设备就可以早于负载过大的网络设备向用户设备返回PADO报文,因此,当用户设备向最早返回PADO报文的网络设备请求建立PPP会话时,用户设备就可以与负载空闲的网络设备建立PPP会话,可见,用户设备能够选择负载空闲的网络设备接入网络,而能够避免选择负载过大的网络设备接入网络,因此,网络中各网络设备的负载更加均衡。
此外,本实施例的技术方案中,控制器可以通过向网络设备下发用于配置第一状态的第一配置信息来实现对网络设备的负载控制。通过这种方式实现网络中各网络设备的负载均衡,一方面用户设备与网络设备之间无需额外增加用于控制报文转发路径的交换设备,因而负载均衡的实现无需改造网络架构,另一方面用户设备上无需预先配置报文转发路径的控制命令,因而负载均衡的实现无需改造大量的用户设备。由此可见,通过本发明实施例的技术方案,网络负载均衡的实现难度小、成本低,因此,网络负载均衡更容易实现。
参见图3,图3为本发明中负载分担的方法另一实施例的流程图。在本实
施例中,所述方法例如具体可以包括如下步骤:
步骤301、目标网络设备接收控制器发送的第一配置信息;
步骤302、所述目标网络设备根据所述第一配置信息,配置第一状态;
所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;
其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
可选的,本实施例还可以包括:
所述目标网络设备接收用户设备发送的PADI报文;
当处于第一状态时,所述目标网络设备延时或者不向所述用户设备返回PADO报文。
可选的,本实施例还可以包括:
所述目标网络设备接收所述控制器发送的负载参数请求;
所述目标网络设备向所述控制器返回所述负载参数。
可选的,本实施例中,所述负载参数请求可以是Openflow协议的扩展的Multipart请求消息,所述负载参数可以通过Openflow协议的的扩展的Multipart应答消息进行发送。
可选的,
所述Mulipart请求消息可以包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;
相应的,所述Mulipart应答消息可以包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,本实施例还可以包括:
所述目标网络设备主动向所述控制器发送所述负载参数。
可选的,本实施例还可以包括:
所述目标网络设备接收所述控制器发送的第二配置信息;
所述目标网络设备根据所述第二配置信息,配置第二状态;
所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;
其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
需要说明的是,本发明实施例中的负载参数、第一配置信息、第二配置信息、第一状态及第二状态的内容,可以参见图2所示实施例。另外,本发明实施例中目标网络设备执行方法步骤的各种具体实施方式,也可以参见图2所示实施例的详细介绍,此处不再赘述。
通过本实施例的技术方案,在网络中,负载过大的网络设备会被控制器配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备延时或拒绝向用户设备返回PADO报文,而不是立即向用户设备返回PADO报文,因此,负载空闲的网络设备就可以早于负载过大的网络设备向用户设备返回PADO报文,因此,当用户设备向最早返回PADO报文的网络设备请求建立PPP会话时,用户设备就可以与负载空闲的网络设备建立PPP会话,可见,用户设备能够选择负载空闲的网络设备接入网络,而能够避免选择负载过大的网络设备接入网络,因此,网络中各网络设备的负载更加均衡。
为了使得本领域技术人员更清楚地理解本发明的具体应用方式,下面以两个示例性的应用场景对本发明实施例进行介绍。
在第一个示例性的应用场景中,第一配置信息为延迟策略,第一状态指示目标网络设备对PADI报文做延时返回PADO报文的处理,第二配置信息为取消延迟策略,第二状态指示目标网络设备对PADI报文做即时返回PADO报文的处理。控制器对第一网络设备和第二网络设备进行负载均衡控制,用户设备需要接入网络时向第一网络设备和第二网络设备同时发起PADI报文。
参见图4,图4为本发明实施例中负载分担的方法又一实施例的流程图。本实施例可以应用到上述第一个示例性的应用场景,所述方法包括以下步骤:
步骤401、控制器获取第一网络设备的第一负载参数及第二网络设备的第二负载参数。
具体实现时,第一网络设备通过Multipart应答消息向控制器发送第一网络设备当前的负载参数,作为第一负载参数;第二网络设备通过Multipart应答消息向控制器发送第二网络设备当前的负载参数,作为第二负载参数。
步骤402、控制器根据第一负载参数确定第一网络设备处于负载超限的状
态,并根据过载设备的记录确定第一网络设备未被记录为过载设备,向第一网络设备发送延迟策略,并将第一网络设备记录为过载设备。
步骤403、第一网络设备配置接收到的延迟策略。
步骤404、控制器根据第二负载参数确定第二网络设备未处于负载超限的状态,并根据过载设备的记录确定第二网络设备未被记录为过载设备,拒绝向第二网络设备发送配置信息。
步骤405、用户设备分别向第一网络设备和第二网络设备发送PADI报文。
步骤406、第二网络设备接收到PADI报文时,即时向用户设备返回PADO报文。
步骤407、第一网络设备接收到PADO报文时,按照已配置的延迟策略,延迟一段时间之后再向用户设备返回PADO报文。
步骤408、用户设备先接收到第二网络设备返回的PADO报文,后接收到第一网络设备返回的PADO报文,用户设备与第二网络设备建立PPP会话。
步骤409、控制器获取第一网络设备的第三负载参数及第二网络设备的第四负载参数。
具体实现时,第一网络设备再次通过Multipart应答消息向控制器发送第一网络设备当前的负载参数,作为第三负载参数;第二网络设备再次通过Multipart应答消息向控制器发送第二网络设备当前的负载参数,作为第四负载参数。
步骤410、控制器根据第三负载参数确定第一网络设备未处于负载超限的状态,并根据过载设备的记录确定第一网络设备已被记录为过载设备,向第一网络设备发送取消延迟策略,并删除第一网络设备作为过载设备的记录。
步骤411、第一网络设备接收到取消延迟策略时,删除延迟策略的配置。
步骤412、控制器根据第四负载参数确定第二网络设备处于负载超限的状态,并根据过载设备的记录确定第二网络设备未被记录为过载设备,向第二网络设备发送延迟策略,并将第二网络设备记录为过载设备。
步骤413、第二网络设备配置接收到的延迟策略。
步骤414、用户设备分别向第一网络设备和第二网络设备发送PADI报文。
步骤415、第一网络设备接收到PADI报文时,即时向用户设备返回PADO报文。
步骤416、第二网络设备接收到PADO报文时,按照已配置的延迟策略,
延迟一段时间之后再向用户设备返回PADO报文。
步骤417、用户设备先接收到第一网络设备返回的PADO报文,后接收到第二网络设备返回的PADO报文,用户设备与第一网络设备建立PPP会话。
通过本实施例的技术方案,通过控制器向网络设备下发延迟策略,负载过大的网络设备对PADI报文做延时返回PADO报文的处理,负载空闲的网络设备对PADI报文做即时返回PADO报文的处理,因此,用户设备能够选择与负载空闲的网络设备建立PPP会话而不是负载过大的网络设备,这样就实现了负载均衡。
在第二个示例性的应用场景中,第一配置信息为阻塞策略,第一状态指示目标网络设备对PADI报文做丢弃处理,第二配置信息为取消阻塞策略,第二状态指示目标网络设备对PADI报文做即时返回PADO报文的处理。控制器对第一网络设备和第二网络设备进行负载均衡控制,用户设备需要接入网络时向第一网络设备和第二网络设备同时发起PADI报文。
参见图5,图5为本发明实施例中负载分担的方法又一实施例的流程图。本实施例可以应用到上述第二个示例性的应用场景,所述方法包括以下步骤:
步骤501、控制器获取第一网络设备的第一负载参数及第二网络设备的第二负载参数。
具体实现时,第一网络设备通过Multipart应答消息向控制器发送第一网络设备当前的负载参数,作为第一负载参数;第二网络设备通过Multipart应答消息向控制器发送第二网络设备当前的负载参数,作为第二负载参数。
步骤502、控制器根据第一负载参数确定第一网络设备处于负载超限的状态,并根据过载设备的记录确定第一网络设备未被记录为过载设备,向第一网络设备发送阻塞策略,并将第一网络设备记录为过载设备。
步骤503、第一网络设备配置接收到的阻塞策略。
步骤504、控制器根据第二负载参数确定第二网络设备未处于负载超限的状态,并根据过载设备的记录确定第二网络设备未被记录为过载设备,拒绝向第二网络设备发送配置信息。
步骤505、用户设备分别向第一网络设备和第二网络设备发送PADI报文。
步骤506、第二网络设备接收到PADI报文时,即时向用户设备返回PADO报文。
步骤507、第一网络设备接收到PADO报文时,按照已配置的阻塞策略,
丢弃PADI报文。
其中,第一网络设备丢弃PADI报文,即第一网络设备拒绝向用户设备返回PADO报文。
步骤508、用户设备接收到第二网络设备返回的PADO报文且未接收到第一网络设备返回的PADO报文,用户设备与第二网络设备建立PPP会话。
步骤509、控制器获取第一网络设备的第三负载参数及第二网络设备的第四负载参数。
具体实现时,第一网络设备再次通过Multipart应答消息向控制器发送第一网络设备当前的负载参数,作为第三负载参数;第二网络设备再次通过Multipart应答消息向控制器发送第二网络设备当前的负载参数,作为第四负载参数。
步骤510、控制器根据第三负载参数确定第一网络设备未处于负载超限的状态,并根据过载设备的记录确定第一网络设备已被记录为过载设备,向第一网络设备发送取消阻塞策略,并删除第一网络设备作为过载设备的记录。
步骤511、第一网络设备接收到取消阻塞策略时,删除阻塞策略的配置。
步骤512、控制器根据第四负载参数确定第二网络设备处于负载超限的状态,并根据过载设备的记录确定第二网络设备未被记录为过载设备,向第二网络设备发送阻塞策略,并将第二网络设备记录为过载设备。
步骤513、第二网络设备配置接收到的阻塞策略。
步骤513、用户设备分别向第一网络设备和第二网络设备发送PADI报文。
步骤514、第一网络设备接收到PADI报文时,即时向用户设备返回PADO报文。
步骤515、第二网络设备接收到PADO报文时,按照已配置的阻塞策略,丢弃PADI报文。
其中,第二网络设备丢弃PADI报文,即第二网络设备拒绝向用户设备返回PADO报文。
步骤515、用户设备接收到第一网络设备返回的PADO报文且未接收到第二网络设备返回的PADO报文,用户设备与第一网络设备建立PPP会话。
通过本实施例的技术方案,通过控制器向网络设备下发阻塞策略,负载过大的网络设备对PADI报文做丢弃处理,负载空闲的网络设备对PADI报文做即时返回PADO报文的处理,因此,用户设备能够选择与负载空闲的网络
设备建立PPP会话而不是负载过大的网络设备,这样就实现了负载均衡。
需要说明的是,上述两个示例性的应用场景仅是本发明实施例的两个示例,本发明实施例不限定于这两个应用场景。
参见图6,图6为本发明中负载分担的装置一实施例的结构示意图。在本实施例中,该装置可以配置于控制器600。该装置包括:
获取模块601,用于获取目标网络设备的负载参数;
第一确定模块602,用于根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;
第一发送模块603,用于当所述目标网络设备处于负载超限的状态时,向所述目标网络设备发送第一配置信息;
所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
具体实现时,在控制器600中,获取模块601获取目标网络设备的负载参数,第一确定模块602根据获取模块601获取到的负载参数确定目标网络设备是否处于负载超限的状态,第一发送模块603在第一确定模块602确定目标网络设备处于负载超限的状态的情况下向目标网络设备发送第一配置信息。
可选的,所述获取模块601可以包括发送子模块和接收子模块。其中,发送子模块,用于向所述目标网络设备发送负载参数请求;接收子模块,用于接收所述目标网络设备对所述负载参数请求而返回的负载参数。
可选的,所述负载参数请求可以是开放流Openflow协议的扩展的多部分Multipart请求消息,所述负载参数可以通过Openflow协议的扩展的Multipart应答消息进行发送。
可选的,所述Mulipart请求消息可以包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;相应的,所述Mulipart应答消息可以包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述获取模块601,可以具体用于接收所述目标网络设备主动发
送的负载参数。
可选的,所述第一确定模块602可以包括查找子模块,计算子模块,第一确定子模块、第二确定子模块和第三确定子模块。其中,查找子模块,用于查找至少两个网络设备的负载参数中的最小值,作为最小负载值;计算子模块,用于计算所述负载参数与所述最小负载值之间的差值,作为负载差值;第一确定子模块,用于确定所述负载差值是否超过预设的负载阈值;第二确定子模块,用于当所述负载差值超过所述负载阈值时,确定所述目标网络设备处于负载超限的状态;第三确定子模块,用于当所述负载差值未超过所述负载阈值时,确定所述目标网络设备未处于负载超限的状态。其中,所述至少两个网络设备中包括所述目标网络设备。
可选的,所述装置还可以包括第二发送模块。其中,第二发送模块,用于当所述所述目标网络设备未处于负载超限的状态时,向所述目标网络设备发送第二配置信息;所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
可选的,所述装置还可以包括记录模块和删除模块。其中,记录模块,用于向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;删除模块,用于向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
可选的,所述装置还可以包括第二确定模块。其中,第二确定模块,用于确定所述目标网络设备是否为已记录的过载设备;所述第一发送模块603,具体用于当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,向所述目标网络设备发送第一配置信息;所述第二发送模块,具体用于当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置信息。
可选的,所述负载参数可以包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
可选的,所述目标网络设备可以为宽带接入网关BNG。
需要说明的是,本实施例中控制器600对应于前述图2所示的实施例中提及的控制器。本实施例中配置于控制器600的装置的各种具体实施方式,
可以参见前述图2所示的实施例的详细介绍,本实施例在此不再赘述。
通过本实施例的技术方案,通过控制器600向负载过大的网络设备下发第一配置信息,负载过大的网络设备被配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备经过控制器的配置将延时或拒绝向用户设备返回PADO报文,而不是立即向用户设备返回PADO报文,因此,用户设备能够选择负载空闲的网络设备接入网络,而能够避免选择负载过大的网络设备接入网络,因此,网络中各网络设备的负载更加均衡。
参见图7,图7为本发明中负载分担的装置另一实施例的结构示意图。在本实施例中,该装置可以配置于目标网络设备700,目标网络设备可以是控制器控制下的任意一个网络设备。该装置包括:
第一接收模块701,用于接收控制器发送的第一配置信息;
第一配置模块702,用于根据所述第一配置信息,配置第一状态;
所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;
其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
具体实现时,在目标网络设备700中,第一接收模块701接收到控制器发送的第一配置信息,第一配置模块702根据第一接收模块701接收到的第一配置信息,配置第一状态。其中,在第一状态的指示下,目标网络设备700接收到PADI时,延时或拒绝返回PADO报文。
可选的,所述装置还可以包括第二接收模块和返回控制模块。其中,第二接收模块,用于接收主机发送的PADI报文;返回控制模块,用于当处于第一状态时,所述目标网络设备延时或者不向所述主机返回PADO报文。
可选的,所述装置还可以包括第三接收模块和返回模块。其中,第三接收模块,用于接收所述控制器发送的负载参数请求;返回模块,用于向所述控制器返回所述负载参数。
可选的,所述负载参数请求可以是Openflow协议的扩展的Multipart请求消息,所述负载参数通过Openflow协议的的扩展的Multipart应答消息进行发送。
可选的,所述Mulipart请求消息可以包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;相应的,所述Mulipart应答消息可以包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述装置还可以包括发送模块。其中,发送模块,用于主动向所述控制器发送所述负载参数。
可选的,所述装置还可以包括第四接收模块和第二配置模块。其中,第四接收模块,用于接收所述控制器发送的第二配置信息;第二配置模块,用于根据所述第二配置信息,配置第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
需要说明的是,本实施例中目标网络设备700对应于前述图2所示的实施例中提及的目标网络设备。本实施例中配置于目标网络设备700的装置的各种具体实施方式,可以参见前述图2所示的实施例的详细介绍,本实施例在此不再赘述。
通过本实施例的技术方案,通过目标网络设备700在负载过大时接收控制器下发的第一配置信息,负载过大的网络设备被配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备经过控制器的配置将延时或拒绝向用户设备返回PADO报文,而不是立即向用户设备返回PADO报文,因此,用户设备能够选择负载空闲的网络设备接入网络,而能够避免选择负载过大的网络设备接入网络,因此,网络中各网络设备的负载更加均衡。
参见图8,图8为本发明中负载分担的***一实施例的结构示意图。在本实施例中,所述***包括控制器801和网络设备802,所述控制器801配置前述图6所示的实施例中任意一种实施方式的装置,所述网络设备802配置有前述图7所示的实施例中任意一种实施方式的装置。
需要说明的是,本实施例中控制器801对应于前述图2所示的实施例中提及的控制器,本实施例中网络设备802对应于前述图2所示的实施例中提及的目标网络设备。本实施例中控制器801和网络设备802的各种具体实施
方式,可以参见前述图2所示的实施例的详细介绍,本实施例在此不再赘述。
通过本实施例的技术方案,控制器将负载过大的网络设备被配置成第一状态,当接收到用户设备发送的PADI时,负载过大的网络设备经过控制器的配置将延时或拒绝向用户设备返回PADO报文,而不是立即向用户设备返回PADO报文,因此,用户设备能够选择负载空闲的网络设备接入网络,而能够避免选择负载过大的网络设备接入网络,因此,网络中各网络设备的负载更加均衡。
此外,网络中各网络设备的负载均衡,既无需在用户设备与网络设备之间额外增加用于控制报文转发路径的交换设备,因而无需改造网络架构,也无需在用户设备上预先配置报文转发路径的控制命令,因而无需改造大量的用户设备,可见,网络负载均衡的实现难度小、成本低,因此,网络负载均衡更容易实现。
参见图9,图9为本发明实施例中一种控制器的结构示意图。在本实施例中,控制器900包括:处理器901、存储器902、网络接口903、总线***904。
所述总线***904,用于将控制器900的各个硬件组件耦合在一起。
所述网络接口903,用于实现控制器900与至少一个其它网元之间的通信连接,可以使用互联网,广域网,本地网,城域网等方式。
所述存储器902,用于存储程序指令和数据。
所述处理器901,用于读取存储器902中存储的指令和数据,执行以下操作:
所述处理器901获取目标网络设备的负载参数;
所述处理器901根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;
当所述目标网络设备处于负载超限的状态时,所述处理器901向所述目标网络设备发送第一配置信息;
所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
可选的,所述处理器901为了获取目标网络设备的负载参数,可以执行以下操作:
所述处理器901向所述目标网络设备发送负载参数请求;
所述处理器901接收所述目标网络设备对所述负载参数请求而返回的负载参数。
可选的,所述负载参数请求可以是开放流Openflow协议的扩展的多部分Multipart请求消息,所述负载参数可以通过Openflow协议的扩展的Multipart应答消息进行发送。
可选的,所述Mulipart请求消息可以包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;相应的,所述Mulipart应答消息可以包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述处理器901为了获取目标网络设备的负载参数,可以执行以下操作:
所述处理器901接收所述目标网络设备主动发送的负载参数。
可选的,所述处理器901为了实现根据所述负载参数确定所述目标网络设备是否处于负载超限的状态,可以执行以下操作:
所述处理器901查找至少两个网络设备的负载参数中的最小值,作为最小负载值;所述至少两个网络设备中包括所述目标网络设备;
所述处理器901计算所述负载参数与所述最小负载值之间的差值,作为负载差值;
所述处理器确901定所述负载差值是否超过预设的负载阈值;
当所述负载差值超过所述负载阈值时,所述处理器901确定所述目标网络设备处于负载超限的状态;
当所述负载差值未超过所述负载阈值时,所述处理器901确定所述目标网络设备未处于负载超限的状态。
可选的,所述处理器901还可以执行以下操作:
当所述目标网络设备未处于负载超限的状态时,所述处理器901向所述目标网络设备发送第二配置信息;
所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
可选的,所述处理器901还可以执行以下操作:
所述处理器901向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;
所述处理器901向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
可选的,所述处理器901还可以执行以下操作:
所述处理器901确定所述目标网络设备是否为已记录的过载设备;
所述处理器901为了实现当所述目标网络设备处于负载超限的状态时向所述目标网络设备发送第一配置信息,具体执行以下操作:当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,所述处理器901向所述目标网络设备发送第一配置信息;
所述处理器为了实现当所述目标网络设备未处于负载超限的状态时向所述目标网络设备发送第二配置信息,具体执行以下操作:当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述处理器901向所述目标网络设备发送第二配置信息。
可选的,所述负载参数可以包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
可选的,所述目标网络设备可以为宽带接入网关BNG。
需要说明的是,本实施例中控制器900对应于前述图2所示的实施例中的控制器。本实施例中控制器900的各种具体实施方式,可以参见前述图2所示的实施例的详细介绍,本实施例在此不再赘述。
参见图10,图10为本发明实施例中一种目标网络设备的结构示意图。在本实施例中,目标网络设备1000包括:处理器1001、存储器1002、网络接口1003、总线***1004。
所述总线***1004,用于将控制器1000的各个硬件组件耦合在一起。
所述网络接口1003,用于实现控制器1000与至少一个其它网元之间的通信连接,可以使用互联网,广域网,本地网,城域网等方式。
所述存储器1002,用于存储程序指令和数据。
所述处理器1001,用于读取存储器1002中存储的指令和数据,执行以下操作:
所述处理器1001接收控制器发送的第一配置信息;
所述处理器1001根据所述第一配置信息,配置第一状态;
所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;
其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
可选的,所述处理器1001还可以执行以下操作:
所述处理器1001接收用户设备发送的PADI报文;
当处于第一状态时,所述处理器1001延时或者不向所述用户设备返回PADO报文。
可选的,所述处理器1001还可以执行以下操作:
所述处理器1001接收所述控制器发送的负载参数请求;
所述处理器1001向所述控制器返回所述负载参数。
可选的,所述负载参数请求可以是Openflow协议的扩展的Multipart请求消息,所述负载参数通过Openflow协议的的扩展的Multipart应答消息进行发送。
可选的,所述Mulipart请求消息可以包括类型Type字段和请求主体Body字段,所述Type字段携带指示负载均衡信息的类型值,所述请求Body字段为空或者携带所述设备标识;相应的,所述Mulipart应答消息可以包括所述Type字段和应答主体Body字段,所述应答Body字段携带所述目标网络设备的负载参数。
可选的,所述处理器1001还可以执行以下操作:
所述处理器1001主动向所述控制器发送所述负载参数。
可选的,所述处理器1001还可以执行以下操作:
所述处理器1001接收所述控制器发送的第二配置信息;
所述处理器1001根据所述第二配置信息,配置第二状态;
所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;
其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
需要说明的是,本实施例中目标网络设备1000对应于前述图2所示的实
施例中的目标网络设备。本实施例中目标网络设备1000的各种具体实施方式,可以参见前述图2所示的实施例的详细介绍,本实施例在此不再赘述。
本发明实施例中提到的“第一配置信息”和“第一发送模块”等名称中的“第一”只是用来做名字标识,并不代表顺序上的第一。该规则同样适用于“第二”、“第三”和“第四”。
需要说明的是,本发明实施例中的处理器可能是一种集成电路芯片,具有信号的处理能力。在实现过程中,上述方法的各步骤可以通过处理器中的硬件的集成逻辑电路或者软件形式的指令完成。这些指令可以通过其中的处理器以配合实现及控制,用于执行本发明实施例揭示的方法。上述处理器还可以是通用处理器、数字信号处理器(Digital Signal Processing,DSP)、专用集成电路(application specific integrated circuit)、现成可编程门阵列(Field Programmable Gate Array,FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件。
其中,上述通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器,解码器等。结合本发明实施例所公开的方法的步骤可以直接体现为硬件处理器执行完成,或者用处理器中的硬件及软件模块组合执行完成。软件模块可以位于随机存储器,闪存、只读存储器,可编程只读存储器或者电可擦写可编程存储器、寄存器等本领域成熟的存储介质中。
另外,需要说明的是,总线***除了包括数据总线之外,还可包括电源总线、控制总线和状态信号总线。但是为了清楚说明起见,在图9、10中将各种总线都标为总线***。
通过以上的实施方式的描述可知,本领域的技术人员可以清楚地了解到上述实施例方法中的全部或部分步骤可借助软件加通用硬件平台的方式来实现。基于这样的理解,本发明的技术方案本质上或者说对现有技术做出贡献的部分可以以软件产品的形式体现出来,该计算机软件产品可以存储在存储介质中,如ROM/RAM、磁碟、光盘等,包括若干指令用以使得一台计算机设备(可以是个人计算机,服务器,或者诸如媒体网关等网络通信设备)执行本发明各个实施例或者实施例的某些部分所述的方法。
需要说明的是,本说明书中的各个实施例均采用递进的方式描述,各个实施例之间相同相似的部分互相参见即可,每个实施例重点说明的都是与其他实施例的不同之处。尤其,对于方法实施例和设备实施例而言,由于其基
本相似于***实施例,所以描述得比较简单,相关之处参见***实施例的部分说明即可。以上所描述的设备及***实施例仅仅是示意性的,其中作为分离部件说明的模块可以是或者也可以不是物理上分开的,作为模块显示的部件可以是或者也可以不是物理模块,即可以位于一个地方,或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部模块来实现本实施例方案的目的。本领域普通技术人员在不付出创造性劳动的情况下,即可以理解并实施。
以上所述仅是本发明的优选实施方式,并非用于限定本发明的保护范围。应当指出,对于本技术领域的普通技术人员来说,在不脱离本发明原理的前提下,还可以作出若干改进和润饰,这些改进和润饰也应视为本发明的保护范围。
Claims (27)
- 一种负载分担的方法,其特征在于,所述方法包括:控制器获取目标网络设备的负载参数;所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息;所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
- 根据权利要求1所述的方法,其特征在于,所述控制器获取目标网络设备的负载参数,包括:所述控制器向所述目标网络设备发送负载参数请求;所述控制器接收所述目标网络设备对所述负载参数请求而返回的负载参数。
- 根据权利要求1所述的方法,其特征在于,所述控制器获取目标网络设备的负载参数,具体为:接收所述目标网络设备主动发送的负载参数。
- 根据权利要求1所述的方法,其特征在于,所述控制器根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态,包括:所述控制器查找至少两个网络设备的负载参数中的最小值,作为最小负载值;所述至少两个网络设备中包括所述目标网络设备;所述控制器计算所述负载参数与所述最小负载值之间的差值,作为负载差值;所述控制器确定所述负载差值是否超过预设的负载阈值;当所述负载差值超过所述负载阈值时,所述控制器确定所述目标网络设备处于负载超限的状态;当所述负载差值未超过所述负载阈值时,所述控制器确定所述目标网络设备未处于负载超限的状态。
- 根据权利要求1所述的方法,其特征在于,还包括:当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标 网络设备发送第二配置信息;所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
- 根据权利要求5所述的方法,其特征在于,还包括:所述控制器向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;所述控制器向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
- 根据权利要求6所示的方法,其特征在于,还包括:所述控制器确定所述目标网络设备是否为已记录的过载设备;所述当所述目标网络设备处于负载超限的状态时,所述控制器向所述目标网络设备发送第一配置信息,具体为:当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,所述控制器向所述目标网络设备发送第一配置信息;所述当所述目标网络设备未处于负载超限的状态时,所述控制器向所述目标网络设备发送第二配置信息,具体为:当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置信息。
- 根据权利要求1所述的方法,其特征在于,所述负载参数包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
- 一种负载分担的方法,其特征在于,所述方法包括:目标网络设备接收控制器发送的第一配置信息;所述目标网络设备根据所述第一配置信息,配置第一状态;所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
- 根据权利要求9所述的方法,其特征在于,还包括:所述目标网络设备接收用户设备发送的PADI报文;当处于第一状态时,所述目标网络设备延时或者不向所述用户设备返回 PADO报文。
- 根据权利要求9所述的方法,其特征在于,还包括:所述目标网络设备接收所述控制器发送的负载参数请求;所述目标网络设备向所述控制器返回所述负载参数。
- 根据权利要求9所述的方法,其特征在于,还包括:所述目标网络设备主动向所述控制器发送所述负载参数。
- 权利要求9所述的方法,其特征在于,还包括:所述目标网络设备接收所述控制器发送的第二配置信息;所述目标网络设备根据所述第二配置信息,配置第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
- 一种负载分担的装置,其特征在于,所述装置包括:获取模块,用于获取目标网络设备的负载参数;第一确定模块,用于根据所述负载参数,确定所述目标网络设备是否处于负载超限的状态;第一发送模块,用于当所述目标网络设备处于负载超限的状态时,向所述目标网络设备发送第一配置信息;所述第一配置信息,用于将所述目标网络设备配置为第一状态;所述第一状态指示所述目标网络设备接收到主动发现发起PADI报文时,延时或者不返回主动发现提供PADO报文。
- 根据权利要求14所述的装置,其特征在于,所述获取模块包括:发送子模块,用于向所述目标网络设备发送负载参数请求;接收子模块,用于接收所述目标网络设备对所述负载参数请求而返回的负载参数。
- 根据权利要求14所述的装置,其特征在于,所述获取模块,具体用于接收所述目标网络设备主动发送的负载参数。
- 根据权利要求14所述的装置,其特征在于,所述第一确定模块包括:查找子模块,用于查找至少两个网络设备的负载参数中的最小值,作为 最小负载值;所述至少两个网络设备中包括所述目标网络设备;计算子模块,用于计算所述负载参数与所述最小负载值之间的差值,作为负载差值;第一确定子模块,用于确定所述负载差值是否超过预设的负载阈值;第二确定子模块,用于当所述负载差值超过所述负载阈值时,确定所述目标网络设备处于负载超限的状态第三确定子模块,用于当所述负载差值未超过所述负载阈值时,确定所述目标网络设备未处于负载超限的状态。
- 根据权利要求14所述的装置,其特征在于,还包括:第二发送模块,用于当所述所述目标网络设备未处于负载超限的状态时,向所述目标网络设备发送第二配置信息;所述第二配置信息,用于将所述目标网络设备配置为第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文。
- 根据权利要求18所述的装置,其特征在于,还包括:记录模块,用于向所述目标网络设备发送第一配置信息之后,将所述目标网络设备记录为过载设备;删除模块,用于向所述目标网络设备发送第二配置信息之后,删除所述目标网络设备作为过载设备的记录。
- 根据权利要求19所述的装置,其特征在于,还包括:第二确定模块,用于确定所述目标网络设备是否为已记录的过载设备;所述第一发送模块,具体用于当所述目标网络设备处于负载超限的状态且所述目标网络设备不为已记录的过载设备时,向所述目标网络设备发送第一配置信息;所述第二发送模块,具体用于当所述目标网络设备未处于负载超限的状态且所述目标网络设备为已记录的过载设备时,所述控制器向所述目标网络设备发送第二配置信息。
- 根据权利要求14所述的装置,其特征在于,所述负载参数包括所述目标网络设备的用户比例、带宽利用率、CPU利用率或剩余CPU处理能力。
- 一种负载分担的装置,其特征在于,所述装置包括:第一接收模块,用于接收控制器发送的第一配置信息;第一配置模块,用于根据所述第一配置信息,配置第一状态;所述第一状态指示所述目标网络设备接收到PADI报文时,延时或者不返回PADO报文;其中,所述第一配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备处于负载超限的状态时所述控制器向所述目标网络设备发送的。
- 根据权利要求22所述的装置,其特征在于,还包括:第二接收模块,用于接收主机发送的PADI报文;返回控制模块,用于当处于第一状态时,所述目标网络设备延时或者不向所述主机返回PADO报文。
- 根据权利要求22所述的装置,其特征在于,还包括:第三接收模块,用于接收所述控制器发送的负载参数请求;返回模块,用于向所述控制器返回所述负载参数。
- 根据权利要求22所述的装置,其特征在于,还包括:发送模块,用于主动向所述控制器发送所述负载参数。
- 根据权利要求22所述的装置,其特征在于,还包括:第四接收模块,用于接收所述控制器发送的第二配置信息;第二配置模块,用于根据所述第二配置信息,配置第二状态;所述第二状态指示所述目标网络设备接收到PADI报文时,即时返回PADO报文;其中,所述第二配置信息是当所述控制器根据所述目标网络设备的负载参数确定所述目标网络设备未处于负载超限的状态时所述控制器向所述目标网络设备发送的。
- 一种负载分担的***,其特征在于,所述***包括控制器和网络设备,所述控制器配置有权利要求14~21任意一项所述的装置,所述网络设备配置有权利要求22~26任意一项所述的装置。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
EP16858878.8A EP3352428B1 (en) | 2015-10-28 | 2016-09-23 | Load sharing method and system |
US15/954,616 US10623320B2 (en) | 2015-10-28 | 2018-04-17 | Load sharing method, apparatus, and system |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201510712755.9 | 2015-10-28 | ||
CN201510712755.9A CN105262698B (zh) | 2015-10-28 | 2015-10-28 | 一种负载分担的方法、装置和*** |
Related Child Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US15/954,616 Continuation US10623320B2 (en) | 2015-10-28 | 2018-04-17 | Load sharing method, apparatus, and system |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2017071442A1 true WO2017071442A1 (zh) | 2017-05-04 |
Family
ID=55102206
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2016/099892 WO2017071442A1 (zh) | 2015-10-28 | 2016-09-23 | 一种负载分担的方法、装置和*** |
Country Status (4)
Country | Link |
---|---|
US (1) | US10623320B2 (zh) |
EP (1) | EP3352428B1 (zh) |
CN (1) | CN105262698B (zh) |
WO (1) | WO2017071442A1 (zh) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105262698B (zh) * | 2015-10-28 | 2019-03-01 | 华为技术有限公司 | 一种负载分担的方法、装置和*** |
CN105763606B (zh) * | 2016-02-04 | 2019-04-09 | 新华三技术有限公司 | 一种服务链代理聚合的方法及*** |
CN105610632B (zh) * | 2016-02-14 | 2019-12-24 | 华为技术有限公司 | 一种虚拟网络设备及相关方法 |
CN107547597A (zh) * | 2016-06-27 | 2018-01-05 | 中兴通讯股份有限公司 | 业务处理方法及装置 |
CN106453124A (zh) * | 2016-10-26 | 2017-02-22 | 杭州华三通信技术有限公司 | 流量调度方法及装置 |
CN106506664B (zh) * | 2016-11-18 | 2020-09-25 | 新华三技术有限公司 | 一种服务器负载均衡的方法和装置 |
CN108541023A (zh) * | 2017-03-03 | 2018-09-14 | 株式会社理光 | 负载均衡方法和***以及接入控制器 |
CN107547321B (zh) * | 2017-06-28 | 2021-05-14 | 新华三技术有限公司 | 报文处理方法、装置、相关电子设备及可读存储介质 |
CN109309627B (zh) * | 2017-07-27 | 2022-05-20 | 中兴通讯股份有限公司 | 负荷分担方法、***及计算机可读存储介质 |
CN108667888B (zh) * | 2017-09-08 | 2020-12-29 | 新华三技术有限公司 | 一种负载均衡方法和装置 |
CN109525536A (zh) * | 2017-09-18 | 2019-03-26 | 深圳市中兴微电子技术有限公司 | 一种ip多媒体子***接入方法和装置 |
US11323320B2 (en) * | 2018-07-27 | 2022-05-03 | Juniper Networks, Inc. | Concurrent transactions on NETCONF devices across network services |
CN112911643B (zh) * | 2019-11-19 | 2023-04-28 | ***通信有限公司政企客户分公司 | 一种负载均衡方法、装置、存储介质和网络设备 |
CN110933182B (zh) * | 2019-12-12 | 2022-06-24 | 新华三大数据技术有限公司 | 负载分担方法、装置及*** |
CN111200649B (zh) * | 2019-12-30 | 2023-03-24 | 北京知道创宇信息技术股份有限公司 | 一种调度方法及装置、调度模型生成方法及装置 |
CN114221965A (zh) * | 2021-12-07 | 2022-03-22 | 杭州趣链科技有限公司 | 访问区块链的方法、装置及电子设备 |
CN115567536B (zh) * | 2022-09-22 | 2024-04-16 | 中国联合网络通信集团有限公司 | 一种算力网络调度方法、装置、电子设备及存储介质 |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20100325295A1 (en) * | 2009-06-17 | 2010-12-23 | Takatoshi Kajiwara | Communication apparatus |
US7984141B2 (en) * | 2007-07-16 | 2011-07-19 | Cisco Technology, Inc. | Independent load balancing for servers |
CN102404794A (zh) * | 2011-11-28 | 2012-04-04 | 福建星网锐捷网络有限公司 | 无线局域网基于频段的负载均衡方法、设备及网络 |
CN104009862A (zh) * | 2013-02-27 | 2014-08-27 | 腾讯科技(深圳)有限公司 | 一种设备调度方法及*** |
CN104102693A (zh) * | 2014-06-19 | 2014-10-15 | 广州华多网络科技有限公司 | 对象处理方法和装置 |
CN105262698A (zh) * | 2015-10-28 | 2016-01-20 | 华为技术有限公司 | 一种负载分担的方法、装置和*** |
Family Cites Families (12)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7487243B1 (en) * | 2003-08-29 | 2009-02-03 | Juniper Networks, Inc. | Network tunnel termination device selection using weighted load balancing |
KR100977901B1 (ko) * | 2004-01-16 | 2010-08-24 | 텔레폰악티에볼라겟엘엠에릭슨(펍) | 교환 이더넷을 통해 지정된 PPPoE 세션 개시 |
CN101087232B (zh) * | 2007-07-27 | 2010-06-09 | 杭州华三通信技术有限公司 | 一种基于以太网上点对点协议的接入方法、***和设备 |
CN102143027A (zh) * | 2010-12-24 | 2011-08-03 | 华为数字技术有限公司 | 接入网络的方法和设备 |
US9071459B2 (en) * | 2012-08-20 | 2015-06-30 | Vonage Business Solutions, Inc. | Methods and systems for establishing electronic communications in a hosted VoIP PBX using two tiered load balancing |
CN104580012B (zh) * | 2013-10-14 | 2018-08-07 | 深圳市腾讯计算机***有限公司 | 一种内容分发网络节点的选择方法及装置 |
CN104883687B (zh) * | 2014-02-28 | 2019-02-26 | 华为技术有限公司 | 无线局域网隧道建立方法、装置及接入网*** |
CN104283804B (zh) * | 2014-10-27 | 2018-05-11 | 新华三技术有限公司 | 一种链路负载均衡方法和装置 |
CN104301247A (zh) * | 2014-10-29 | 2015-01-21 | 杭州华三通信技术有限公司 | 一种pppoe接入设备负载均衡的方法和装置 |
US10592379B2 (en) * | 2014-11-18 | 2020-03-17 | Comcast Cable Communications Management, Llc | Methods and systems for status determination |
CN104836746B (zh) * | 2015-05-08 | 2018-11-27 | 新华三技术有限公司 | PPPoE网络报文转发的方法及装置 |
CN105262696B (zh) | 2015-09-01 | 2019-02-12 | 上海华为技术有限公司 | 一种多路径分流的方法及相关设备 |
-
2015
- 2015-10-28 CN CN201510712755.9A patent/CN105262698B/zh active Active
-
2016
- 2016-09-23 WO PCT/CN2016/099892 patent/WO2017071442A1/zh unknown
- 2016-09-23 EP EP16858878.8A patent/EP3352428B1/en active Active
-
2018
- 2018-04-17 US US15/954,616 patent/US10623320B2/en not_active Expired - Fee Related
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7984141B2 (en) * | 2007-07-16 | 2011-07-19 | Cisco Technology, Inc. | Independent load balancing for servers |
US20100325295A1 (en) * | 2009-06-17 | 2010-12-23 | Takatoshi Kajiwara | Communication apparatus |
CN102404794A (zh) * | 2011-11-28 | 2012-04-04 | 福建星网锐捷网络有限公司 | 无线局域网基于频段的负载均衡方法、设备及网络 |
CN104009862A (zh) * | 2013-02-27 | 2014-08-27 | 腾讯科技(深圳)有限公司 | 一种设备调度方法及*** |
CN104102693A (zh) * | 2014-06-19 | 2014-10-15 | 广州华多网络科技有限公司 | 对象处理方法和装置 |
CN105262698A (zh) * | 2015-10-28 | 2016-01-20 | 华为技术有限公司 | 一种负载分担的方法、装置和*** |
Also Published As
Publication number | Publication date |
---|---|
EP3352428A1 (en) | 2018-07-25 |
EP3352428A4 (en) | 2018-08-29 |
US10623320B2 (en) | 2020-04-14 |
US20180234344A1 (en) | 2018-08-16 |
EP3352428B1 (en) | 2020-11-11 |
CN105262698A (zh) | 2016-01-20 |
CN105262698B (zh) | 2019-03-01 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2017071442A1 (zh) | 一种负载分担的方法、装置和*** | |
US11576114B2 (en) | Data packet processing method, network traffic management method, apparatus, and system | |
US10251091B2 (en) | Traffic offloading method and apparatus | |
US9641650B2 (en) | TCP proxy server | |
US10594606B2 (en) | Wired data-connection aggregation | |
US8806030B2 (en) | Multichannel connections in file system sessions | |
US10461986B2 (en) | Data transmission method, apparatus, and system | |
JP2018508151A (ja) | 伝送制御プロトコルtcpデータパケットを送信する方法及び装置、並びにシステム | |
JP5923376B2 (ja) | Tcp中継装置 | |
WO2015096149A1 (zh) | Tcp链路配置方法、装置及设备 | |
US20210329528A1 (en) | Connection establishment method and related device | |
WO2018077100A1 (zh) | Tcp拥塞窗口的确定方法和装置 | |
US11503498B2 (en) | Information-centric networking over 5G or later networks | |
WO2023035925A1 (zh) | 一种业务处理方法、装置和*** | |
WO2015074537A1 (zh) | 一种智能电视设备中的通信协议控制方法和装置 | |
EP4011044A1 (en) | Technique for controlling and performing data traffic handling in a core network domain | |
RU2589867C2 (ru) | Устройство связи, устройство управления, система связи, способ связи, способ управления устройством связи и программа | |
CN104521201B (zh) | 转发节点的处理方法、转发节点及控制节点 | |
CN110391991B (zh) | 一种流量控制的方法及相关装置 | |
JP5162641B2 (ja) | メディアアダプタ装置とルータ装置の自動設定システム | |
Zhou et al. | Janus: A multi-TCP framework for application-aware optimization in mobile networks | |
JP7217647B2 (ja) | 通信装置、通信方法、及びプログラム | |
JP2013038485A (ja) | 信号処理装置および信号処理方法 |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
121 | Ep: the epo has been informed by wipo that ep was designated in this application |
Ref document number: 16858878 Country of ref document: EP Kind code of ref document: A1 |
|
NENP | Non-entry into the national phase |
Ref country code: DE |