WO2017113346A1 - 负荷分担的方法和业务交换机 - Google Patents

负荷分担的方法和业务交换机 Download PDF

Info

Publication number
WO2017113346A1
WO2017113346A1 PCT/CN2015/100237 CN2015100237W WO2017113346A1 WO 2017113346 A1 WO2017113346 A1 WO 2017113346A1 CN 2015100237 W CN2015100237 W CN 2015100237W WO 2017113346 A1 WO2017113346 A1 WO 2017113346A1
Authority
WO
WIPO (PCT)
Prior art keywords
service
virtual machine
service data
application service
data flow
Prior art date
Application number
PCT/CN2015/100237
Other languages
English (en)
French (fr)
Inventor
韩霆军
李晓
李力军
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2015/100237 priority Critical patent/WO2017113346A1/zh
Publication of WO2017113346A1 publication Critical patent/WO2017113346A1/zh

Links

Images

Definitions

  • the embodiments of the present invention relate to the field of communications technologies, and in particular, to a load balancing method and a service switch.
  • FIG. 1 is a schematic diagram of a network architecture of a service data flow of a user equipment (User Equipment, UE for short) accessing a network in the prior art. As shown in FIG. 1 , UE A and UE B access the network (English: Internet). The service data flows through the classifier (English: Classifier) and needs to be processed by the same service chain after being diverted according to the user, service, and network status.
  • SFC Service Function Chain
  • two application services are integrated, namely, application service 1 and application service 2, wherein the application services are: security charging, security protection, content optimization, content filtering, caching service, firewall, etc.;
  • the service data flows of the UE A and the UE B accessing the network are processed by the application service, that is, the service data flows of the UE A and the UE B accessing the network are processed by the application service 1 and then processed by the application service 2;
  • the processing sequence of processing the application service 1 and then processing the application service 2 is called a logical chain and can be expressed as (application service 1 -> application service 2).
  • the application processing service 1 has two instances, and the application service 2 instance (that is, the virtual machine) is two, which are respectively instance 1 and instance 2, and the data is processed by which instance of each application service.
  • the processing sequence is called a physical chain, and the instance refers to a virtual machine for providing application services. Therefore, the process of determining which instance of each application service to perform application service processing on the service data flow is: for each application service, the hash value is calculated according to the IP address of the UE, and then the processing application is determined according to the hash value. An instance of the service is processed by the service. For the same UE, the hash value calculated for each application service is the same. For example, for application service 1, the hash value is calculated according to the IP address of UE A.
  • the instance 1 of the application service 1 is selected to perform the processing of the application service 1 on the service data flow of the UE A;
  • the IP address of the UE A is calculated to have a hash value of 1.
  • the instance 1 of the application service 2 is selected to perform the processing of the application service 2 on the service data stream of the UE B. Therefore, the physical chain of the UE A is (application service 1: Example 1 -> Application Service 2: Example 1).
  • the hash value is calculated according to the IP address of the UE B.
  • the instance 2 of the application service 1 is selected to perform the processing of the application service 1 on the service data flow of the UE B.
  • the IP address of the UE B is calculated to have a hash value of 2.
  • the instance 2 of the application service 2 is selected to perform the processing of the application service 2 on the service data stream of the UE B. Therefore, the physical chain of the UE B is (application service 1: Example 2 -> Application Service 2: Example 2).
  • the instance 1 of the application service 1 and the instance 2 of the application service 2 process the service data flow of the UE, or the instance 2 of the application service 1 and the instance 2 of the application service 2 are processed by the UE.
  • the service data stream is processed; if any of the above instances fails, the probability that the UE's service data flow is half is not processed; and if the instance 1 of the application service 1 and the instance 2 of the application service 2 are both failed,
  • the service data flows of all UEs cannot be processed, causing all application service processing to be interrupted; therefore, the reliability of the entire network system is reduced.
  • the embodiment of the present invention provides a load balancing method and a service switch, which are used to reduce the risk that the service data flow of the UE cannot be processed by the application service, avoid interruption of all application service processing, and improve the reliability of the entire network system. .
  • the embodiment of the present invention provides a load balancing method, including: a service switch receives a service data flow of a UE; and provides the first application according to the identifier information of the UE and the indication information of the first application service. Determining a first virtual machine in at least two virtual machines of the service; and then sending the service data flow of the UE to the first virtual machine, so that the first virtual machine performs a service data flow to the UE The processing of the first application service is performed; and the service data flow of the UE processed by the first virtual machine is received; and the service data flow of the UE processed by the first virtual machine is sent to the target device.
  • the virtual machine for processing the service data flow of the UE in the embodiment is determined by the identification information of the UE and the indication information of the application service provided by the virtual machine, the virtual machine of one application service may be combined with other applications.
  • Each virtual machine of the service belongs to a different physical chain that processes the service data stream, and if one of the virtual machines in the physical chain thus formed A failure occurs, and other virtual machines in the physical chain may also belong to other physical chains, which avoids the problem that other virtual machines cannot process the service data flow; and as long as there is a normal virtual machine in each application service, these normal virtual machines
  • the opportunity forms a physical chain for processing the service data flow, thereby reducing the risk that the service data flow cannot be processed by the application service, avoiding the interruption of all application service processing, and improving the reliability of the entire network system.
  • the indication information is an identifier of the first application service
  • the indication information is an identifier of the service switch or a random value of the service switch.
  • the identifier information of the UE is an IP address or a device number of the UE.
  • the service switch is The identifier information and the indication information of the first application service determine the first virtual machine from the at least two virtual machines that provide the first application service, including:
  • the service switch performs hash operation on the identifier information of the UE and the indication information to obtain a hash value
  • the service switch determines, according to the hash value, the first virtual machine corresponding to the hash value from at least two virtual machines that provide the first application service.
  • the target device is the UE or an Internet server.
  • the service switch sends the service data flow of the UE to the first Before the virtual machine, it also includes:
  • the service switch Determining, by the service switch, the second virtual machine from at least two virtual machines that provide processing of the second application service, according to the identifier information of the UE and the indication information of the second application service;
  • the service switch sends the service data stream of the UE to the second virtual machine, so that the second virtual machine performs the processing of the second application service on the service data flow of the UE;
  • the service switch sends the service data stream of the UE to the first virtual machine, where the service switch sends the service data stream of the UE processed by the second virtual machine to the service
  • the first virtual machine is described.
  • the embodiment of the present invention provides a service switch, including: a receiving unit, configured to receive a service data flow of the UE, and a determining unit, configured to: according to the identifier information of the UE and the indication information of the first application service, Determining a first virtual machine from the at least two virtual machines that provide the first application service, and sending, by the sending unit, the service data flow of the UE to the first virtual machine, to enable the first virtual machine Performing processing of the first application service on the service data flow of the UE; the receiving unit is further configured to receive a service data flow of the UE processed by the first virtual machine; And sending, to the target device, the service data flow of the UE processed by the first virtual machine.
  • the virtual machine for processing the service data flow of the UE in the embodiment is determined by the identification information of the UE and the indication information of the application service provided by the virtual machine, the virtual machine of one application service may be combined with other applications.
  • Each virtual machine of the service belongs to a different physical chain that processes the service data stream. If one of the virtual machines in the physical chain thus formed fails, and other virtual machines in the physical chain may also belong to other physical chains, the avoidance is avoided.
  • Other virtual machines cannot handle the problem of service data flow; and as long as there is a normal virtual machine in each application service, these normal virtual opportunities form a physical chain that processes the service data flow, thereby reducing the service data flow that cannot be processed by the application service. Risk, avoiding the interruption of all application business processing, and improving the reliability of the entire network system.
  • the indication information is an identifier of the first application service
  • the indication information is an identifier of the service switch or a random value of the service switch.
  • the identifier information of the UE is an IP address or a device number of the UE.
  • the determining unit is specifically used to And performing a hash operation on the identifier information of the UE and the indication information to obtain a hash value; and determining, according to the hash value, from the at least two virtual machines that provide the first application service The first virtual machine corresponding to the hash value.
  • the target device is the UE or an Internet server.
  • the determining unit is further configured to: before the sending unit sends the service data stream of the UE to the first virtual machine, according to the identifier information of the UE, and the second Determining, by the indication information of the service, the second virtual machine from at least two virtual machines that provide processing of the second application service;
  • the sending unit is further configured to send the service data flow of the UE to the second virtual machine, so that the second virtual machine performs the processing of the second application service on the service data flow of the UE. ;
  • the receiving unit is further configured to receive a service data flow of the UE processed by the second virtual machine
  • the sending unit When the sending unit sends the service data stream of the UE to the first virtual machine, the sending unit is configured to send the service data stream of the UE processed by the second virtual machine to the first virtual machine.
  • an embodiment of the present invention provides a service switch, including: a memory, a processor, and a communication interface;
  • the memory configured to store program code for performing a method of load sharing
  • the processor is configured to invoke the program code execution in the memory:
  • the first virtual machine Determining, according to the identifier information of the UE and the indication information of the first application service, the first virtual machine from the at least two virtual machines that provide the first application service;
  • the indication information is an identifier of the first application service
  • the indication information is an identifier of the service switch or a random value of the service switch.
  • the identifier information of the UE is an IP address or a device number of the UE.
  • the processor is at the root Determining, according to the identifier information of the UE and the indication information of the first application service, the first virtual machine from the at least two virtual machines that provide the first application service, specifically: the identifier information of the UE and Determining, by the hash operation, a hash value, and determining, according to the hash value, the first one corresponding to the hash value from at least two virtual machines providing the first application service virtual machine.
  • the target device is the UE or an Internet server.
  • the processor is further configured to: Before the service data stream is sent to the first virtual machine, determining, according to the identifier information of the UE and the indication information of the second application service, from the at least two virtual machines that provide the processing of the second application service.
  • the second virtual machine sends the service data stream of the UE to the second virtual machine by using the communication interface, so that the second virtual machine performs the second application service on the service data flow of the UE Receiving, by the communication interface, the service data flow of the UE processed by the second virtual machine;
  • the processor When the processor sends the service data stream of the UE to the first virtual machine by using the communication interface, the processor is specifically configured to: process the UE processed by the second virtual machine by using the communication interface The service data stream is sent to the first virtual machine.
  • the virtual machine for processing the service data flow of the UE in the embodiment is determined by the identifier information of the UE and the indication information of the application service provided by the virtual machine,
  • a virtual machine of an application service may belong to a different physical chain of a service data flow separately from each virtual machine of another application service, and if one of the virtual machines in the thus formed physical network fails, and other virtual machines in the physical chain It may also belong to other physical chains at the same time, avoiding the problem that other virtual machines cannot handle the service data flow; and as long as there is a normal virtual machine in each application service, these normal virtual opportunities form a physical chain for processing the service data stream, thereby Reduce the risk that the service data flow cannot be processed by the application service, avoid the interruption of all application service processing, and improve the reliability of the entire network system.
  • FIG. 1 is a schematic diagram of a network architecture of a service data flow of a UE accessing a network in the prior art
  • FIG. 2 is a flowchart of a method for load sharing according to Embodiment 1 of the present invention.
  • FIG. 3 is a flowchart of a method for load sharing according to Embodiment 2 of the present invention.
  • FIG. 4 is a schematic diagram of forming a physical chain according to an embodiment of the present invention.
  • FIG. 5 is a schematic structural diagram of a service switch according to Embodiment 1 of the present invention.
  • FIG. 6 is a schematic structural diagram of a service switch according to Embodiment 2 of the present invention.
  • the service switch 1 shown in FIG. 1 can implement the implementation solution provided by the embodiments of the present invention, as shown in FIG.
  • the switch 2 can also implement the implementation provided by the embodiments of the present invention.
  • FIG. 2 is a flowchart of a method for load sharing according to Embodiment 1 of the present invention. As shown in FIG. 2, the method in this embodiment may include:
  • the service switch receives a service data flow of the UE.
  • the service switch determines the first virtual machine from the at least two virtual machines that provide the first application service according to the identifier information of the UE and the indication information of the first application service.
  • the service switch may determine that the service data of the UE accessing the network needs to be processed by the first application service, where the first application service may need to process the service data.
  • the service switch determines, according to the identifier information of the UE and the indication information of the first application service, the virtual service for processing the first application service of the service data flow of the UE from the at least two virtual machines that provide the first application service.
  • the virtual machine is referred to as a first virtual machine; the first virtual machine is a virtual machine on a physical chain for processing a service data flow of the UE, and the physical chain refers to each application service that sequentially processes the service data flow of the UE.
  • the sequence of virtual machines is referred to as a first virtual machine; the first virtual machine is a virtual machine on a physical chain for processing a service data flow of the UE, and the physical chain refers to each application service that sequentially processes the service data flow of the UE. The sequence of virtual machines.
  • the service data flow of the UE received by the service switch may be sent by a previous device of the service switch in the flow direction of the service data flow, or sent by a virtual machine that provides the second application service.
  • the application service is the last application service of the first application service in the sequence of applying the service processing to the service data stream of the UE.
  • the identifier information of the UE is an IP address or a device number of the UE, or other information used to uniquely identify the UE.
  • the service switch sends the service data stream of the UE to the first virtual machine, so that the first virtual machine performs the processing of the first application service on the service data flow of the UE.
  • the service switch receives the service data flow of the UE processed by the first virtual machine.
  • the service switch sends the service data stream of the UE processed by the first virtual machine to the target device.
  • the service switch after determining the first virtual machine, sends the received service data flow of the UE to the first virtual machine, and the first virtual machine performs the first application service on the service data flow of the UE.
  • the first virtual machine sends the service data stream of the UE processed by the first virtual machine to the service switch, and after receiving the service data flow of the UE processed by the first virtual machine, the service switch The service data stream of the UE processed by the first virtual machine is sent to the target device.
  • the target device is an Internet server, that is, the service switch sends the service data flow of the processed UE through the application service to the Internet server. If the service data flow of the UE is sent from the Internet to the UE, the target device is the UE, that is, the service switch sends the service data flow of the processed UE through the application service to the UE.
  • a specific implementation manner of the S102 is: the service switch performs hash operation on the identifier information of the UE and the indication information of the first application service, obtains a hash value, and then according to the hash. And determining, by the at least two virtual machines providing the first application service, the first virtual machine corresponding to the hash value.
  • the hash values obtained by the operation may be the same or different; thus, a virtual machine corresponding to a hash value in one application service and other application services
  • the virtual machines corresponding to the different hash values in the UE jointly perform the application service processing on the service data flow of the UE, and may also perform the service data flow of the UE together with the virtual machine corresponding to the same hash value in the other application services.
  • the load balancing method provided in this embodiment is determined by the identifier information of the UE and the indication information of the application service provided by the virtual machine, because the virtual machine for processing the service data flow of the UE in the embodiment is determined by
  • a virtual machine of an application service may belong to a different physical chain of a service data flow separately from each virtual machine of another application service, and if one of the virtual machines in the thus formed physical network fails, and other virtual machines in the physical chain It may also belong to other physical chains at the same time, avoiding the problem that other virtual machines cannot handle the service data flow; and as long as there is a normal virtual machine in each application service, these normal virtual opportunities form a physical chain for processing the service data stream, thereby Reduce the risk that the service data flow cannot be processed by the application service, avoid the interruption of all application service processing, and improve the reliability of the entire network system.
  • FIG. 3 is a flowchart of a method for load sharing according to Embodiment 2 of the present invention. As shown in FIG. 3, the method in this embodiment may include:
  • the service switch receives the first service data flow of the UE.
  • the service switch determines the second virtual machine from the at least two virtual machines that provide the processing of the second application service according to the identifier information of the UE and the indication information of the second application service.
  • the service switch sends the first service data flow of the UE to the second virtual machine.
  • the service switch may determine that the service data flow of the UE needs to be processed by the second application service before the processing of the first application service.
  • the service switch receives the first service data stream of the UE, and then determines a virtual machine from the at least two virtual machines that provide the processing of the second application service according to the identifier information of the UE and the indication information of the second application service, where the virtual machine is The second virtual machine. Then, the service switch sends the first service data stream of the UE to the second virtual machine, so that the second virtual machine performs the processing of the second application service on the first service data stream of the UE.
  • the second virtual machine After receiving the first service data flow of the UE, the second virtual machine performs processing of the second application service on the first service data flow of the UE, and obtains the second service data flow of the UE, and the second service data of the UE Flowing is the first service data flow of the UE processed by the second virtual machine; the second virtual machine is the second The service data stream is sent to the service switch.
  • the first service data flow may be sent to the service switch by the last hop device of the service switch, or the third virtual machine may send the service data flow of the UE processed by the third virtual machine to the service
  • the third virtual machine is a virtual machine that provides a third application service
  • the third application service is a previous application service of the second application service in a sequence of application service processing on the service data flow of the UE.
  • the service switch determines the implementation scheme of the second virtual machine from the at least two virtual machines that provide the processing of the second application service according to the identifier information of the UE and the indication information of the second application service, and the service switch according to the UE
  • the identifier information and the indication information of the first application service are similar to the implementation of determining the first virtual machine from the at least two virtual machines that provide the processing of the first application service, and details are not described herein again.
  • the service switch receives the second service data flow of the UE sent by the second virtual machine.
  • the service switch receives the second service data flow of the UE that is sent by the second virtual machine.
  • the service switch determines the first virtual machine from the at least two virtual machines that provide the first application service according to the identifier information of the UE and the indication information of the first application service.
  • S205 may be executed before S206. This embodiment does not limit S205 to be executed after S204.
  • the service switch sends the second service data flow of the UE to the first virtual machine.
  • the service switch receives a third service data flow of the UE that is sent by the first virtual machine.
  • the service switch after determining the first virtual machine, sends the second service data stream (that is, the first service data stream processed by the second virtual machine) to the first virtual machine, where the first virtual machine receives the first
  • the second service data flow is processed by the first service data flow
  • the third service data flow is the second service data flow processed by the first virtual machine, that is, the first service data flow.
  • the service switch sends the third service data flow of the UE to the target device.
  • the foregoing solution can reduce that the service data flow cannot be processed by the application service. Risk, avoiding the interruption of all application business processing, and improving the reliability of the entire network system.
  • UE1, UE2, UE3, and UE4 all go through the same service chain, and the application service processing order of the service data flows of the four UEs is: first apply service 1 and then apply service 2; wherein, provide virtual application service 1
  • the virtual machine 1 of the application service 1 and the virtual machine 2 of the application service 1 include: the virtual machine 1 of the application service 1 and the virtual machine 2 of the application service 2, if the service to each UE
  • the data flow is processed by the application service 1 and the processing of the application service 2 is forwarded by the same switch.
  • the indication of the application service is used as an example of the identifier of the application service.
  • the service switch determines that the virtual machine 1 of the application service 1 processes the service data flow of the UE1 according to the IP address of the UE1 and the identifier of the application service 1, and the service switch is further configured according to the UE1.
  • the IP address and the identifier of the application service 2 determine that the virtual machine 1 of the application service 2 performs the processing of the application service 2 on the service data flow of the UE1; therefore, each virtual machine that processes the service data flow of the UE1 for each application service
  • the order is: first apply the virtual machine 1 of the service 1 and then apply the virtual machine 1 of the service 2. For example, as shown in FIG.
  • the service switch performs hash operation on the IP address of UE1 as 10.1.1.1 and the identifier of the application service 1 as 1.
  • the obtained hash value is 1, and the virtual machine corresponding to the hash value is 1.
  • Applying the virtual machine 1 of the service 1 the service switch determines the virtual machine 1 of the application service 1 from the virtual machine providing the first application service; the service switch performs the IP address of the UE1 as 10.1.1.1 and the identifier of the application service 2 as 2.
  • the hash operation obtains a hash value of 1, and the virtual machine corresponding to the hash value of 1 is the virtual machine 1 of the application service 1; therefore, the service data flow of the UE1 is first processed by the virtual machine 1 of the application service 1 and then by the application.
  • the sequence processed by the virtual machine 1 of the service 2 is referred to as the physical chain 1 as shown in FIG.
  • the service switch determines that the virtual machine 1 of the application service 1 processes the service data flow of the UE2 according to the IP address of the UE2 and the identifier of the application service 1, and the service switch further according to the UE2.
  • the IP address and the identifier of the application service 2, and the virtual machine 2 of the application service 2 determines that the service data flow of the UE2 is processed by the application service 2; therefore, each virtual machine of the service data flow of the UE2 is processed by each application service.
  • the order is: first apply the virtual machine 1 of the service 1 and then apply the virtual machine 2 of the service 2.
  • the service switch performs the hash operation with the IP address of the UE2 being 10.1.1.2 and the identifier of the application service 1 being 1.
  • the obtained hash value is 1, and the virtual machine corresponding to the hash value of 1 is the virtual machine 1 of the first application service. Therefore, the service switch determines the virtual machine 1 of the application service 1 from the virtual machine providing the application service 1; the service switch performs the hash operation on the IP address of the UE 2 as 10.1.1.2 and the identifier of the application service 2, and obtains the hash.
  • the value is 2, and the virtual machine corresponding to the hash value of 2 is the virtual machine 2 of the application service 2; therefore, the service data flow of the UE 2 is first processed by the virtual machine 1 of the application service 1 and then processed by the virtual machine 2 of the application service 2.
  • the sequence is referred to as physical chain 2 as shown in FIG.
  • the service switch determines that the virtual machine 2 of the application service 1 processes the service data flow of the UE3 according to the IP address of the UE3 and the identifier of the application service 1, and the service switch further performs the service according to the UE3.
  • the IP address and the identifier of the application service 2, and the virtual machine 2 of the application service 2 determines that the service data flow of the UE3 is processed by the application service 2; therefore, each virtual machine that processes the service data flow of the UE3 is processed by each application service.
  • the order is: first apply the virtual machine 2 of the service 1 and then apply the virtual machine 2 of the service 2. For example, as shown in FIG.
  • the service switch performs hash operation on the IP address of UE3 as 10.1.1.3 and the identifier of the application service 1 as 1.
  • the obtained hash value is 2, and the virtual machine corresponding to the hash value of 2 is Applying the virtual machine 2 of the service 1, the service switch determines the virtual machine 2 of the application service 1 from the virtual machine providing the application service 1.
  • the service switch sets the IP address of the UE3 to 10.1.1.3 and the identifier of the application service 2 to 2.
  • the hash operation has a hash value of 2, and the virtual machine corresponding to the hash value of 2 is the virtual machine 2 of the application service 2; therefore, the service data flow of the UE3 is first processed by the virtual machine 2 of the application service 1 and then by the application service.
  • the sequence processed by the virtual machine 2 of 2 is referred to as the physical chain 3 as shown in FIG.
  • the service switch determines that the virtual machine 2 of the application service 1 processes the service data flow of the UE4 according to the IP address of the UE4 and the identifier of the application service 1, and the service switch further according to the UE4.
  • the IP address and the identifier of the application service 2 determine that the virtual machine 1 of the application service 2 performs the processing of the application service 2 on the service data flow of the UE 4; therefore, the virtual machine of the service data flow of the UE 4 performs the processing of each application service.
  • the order is: first apply the virtual machine 2 of the service 1 and then apply the virtual machine 1 of the service 2. For example, as shown in FIG.
  • the service switch performs the hash operation on the IP address of UE4 as 10.1.1.4 and the identifier of the application service 1 as 1, and obtains a hash value of 2, and the virtual machine corresponding to the hash value of 2 is Applying the virtual machine 2 of the service 1, the service switch determines the virtual machine 2 of the application service 1 from the virtual machine providing the application service 1. The service switch sets the IP address of the UE4 to 10.1.1.4 and the identifier of the application service 2 to 2.
  • service switch of the virtual machine that determines the processing of the application service 1 for the service number flow of each UE and the service switch of the virtual machine that performs the processing of the application service 2 for the service data flow of each UE may not be the same. Service switch.
  • the service switch determines the virtual machine that performs the processing of the application service according to the IP address of the UE and the identifier of the application service, and the physical chain of the service data flow of the processed UE formed by the service is multiple, so as shown in FIG.
  • the virtual machine 1 of the application service 1 has failed, and the virtual machine 2 of the application service 2 and the virtual machine 2 of the application service 2 respectively form a physical chain for processing the service data flow with the virtual machine 2 of the application service 1, and thus will not
  • Each virtual machine that affects the application service 2 performs processing of the application service 2 on the service data stream.
  • the service switch determines the virtual machine that provides the application service according to the IP address of the UE, for the same UE, it is determined that the virtual machine 1 and the application service 2 of the application service 1 are virtual.
  • the machine 1 is either a virtual machine 2 that determines the application service 1 and a virtual machine 2 of the application service 2; a total of two physical chains that process the service data flow of the UE are formed: one physical chain is: the virtual machine 1 that applies the service 1 first.
  • the virtual machine 1 of the service 2 is applied again, and the other physical chain is: the virtual machine 2 of the service 1 is applied first, and the virtual machine 2 of the service 2 is applied; if the virtual machine 1 of the application service 1 and the virtual machine 2 of the application service 2 appear If the fault occurs, the two physical chains are unavailable, which causes all application service processing to be interrupted, which affects the reliability of the entire network system.
  • the indication information of the application service may be an identifier of the service switch or a random value of the service switch, and the identifier of the service switch is used to distinguish it from other service switches, and the specific implementation process
  • a service switch refer to the identifier of the application service to determine related records in the virtual machine of the application service, and details are not described herein.
  • the random value of the service switch is a value randomly generated by the service switch deployed in the network. Values are used to distinguish them from other service switches.
  • FIG. 5 is a schematic structural diagram of a service switch according to Embodiment 1 of the present invention.
  • the service switch in this embodiment may include: a receiving unit 11, a determining unit 12, and a sending unit 13.
  • the receiving unit 11 is configured to receive the service data stream of the UE
  • the determining unit 12 is configured to provide at least two virtualities of the first application service according to the identifier information of the UE and the indication information of the first application service.
  • the sending unit 13 is configured to send the service data flow of the UE to the first virtual machine, so that the first virtual machine performs the service data flow to the UE a processing of the first application service, the receiving unit 11 is further configured to receive the service data flow of the UE processed by the first virtual machine, and the sending unit 13 is further configured to: process the first virtual machine The service data stream of the UE is sent to the target device.
  • the indication information is an identifier of the first application service
  • the indication information is an identifier of the service switch or a random value of the service switch.
  • the identifier information of the UE is an IP address or a device number of the UE.
  • the determining unit 12 is specifically configured to: perform hash operation on the identifier information of the UE and the indication information, obtain a hash value, and provide the first application service according to the hash value.
  • the first virtual machine corresponding to the hash value is determined in at least two virtual machines.
  • the target device is the UE or an Internet server.
  • the determining unit 12 is further configured to provide, according to the identifier information of the UE and the indication information of the second application service, before the sending unit 13 sends the service data stream of the UE to the first virtual machine. Determining, by the at least two virtual machines of the processing of the second application service, the second virtual machine;
  • the sending unit 13 is further configured to send the service data stream of the UE to the second virtual machine, so that the second virtual machine performs the processing of the second application service on the service data flow of the UE;
  • the receiving unit 11 is further configured to receive a service data flow of the UE processed by the second virtual machine;
  • the sending unit 13 is configured to: when the service data stream of the UE is sent to the first virtual machine, send the service data stream of the UE processed by the second virtual machine to the first virtual machine.
  • the service switch of this embodiment may be used to implement the technical solutions of the foregoing method embodiments of the present invention, and the implementation principles and technical effects thereof are similar, and details are not described herein again.
  • FIG. 6 is a schematic structural diagram of a service switch according to Embodiment 2 of the present invention.
  • the service switch in this embodiment may include: a memory 21, a processor 22, and a communication interface 23; wherein, the memory 21 is configured to store a program code for performing a load sharing method; the processor 22, configured to invoke the program code execution in the memory 21:
  • the first virtual machine Determining, according to the identifier information of the UE and the indication information of the first application service, the first virtual machine from the at least two virtual machines that provide the first application service;
  • the service data stream of the UE processed by the first virtual machine is sent to the target device through the communication interface 23.
  • the indication information is an identifier of the first application service
  • the indication information is an identifier of the service switch or a random value of the service switch.
  • the identifier information of the UE is an IP address or a device number of the UE.
  • the processor 22 is configured to: when determining the first virtual machine from the at least two virtual machines that provide the first application service, according to the identifier information of the UE and the indication information of the first application service, specifically: Performing a hash operation on the identifier information of the UE and the indication information to obtain a hash value; and determining, according to the hash value, from the at least two virtual machines that provide the first application service The first virtual machine corresponding to the hash value.
  • the target device is the UE or an Internet server.
  • the processor 22 is further configured to: before sending the service data flow of the UE to the first virtual machine by using the communication interface 23, according to the identifier information of the UE and the indication information of the second application service, Determining the second virtual machine from at least two virtual machines providing processing of the second application service; transmitting, by the communication interface 23, the service data stream of the UE to the second virtual machine, so that the The second virtual machine performs the processing of the second application service on the service data flow of the UE; and receives the service data flow of the UE processed by the second virtual machine by using the communication interface 23;
  • the processor 22 When the processor 22 sends the service data stream of the UE to the first virtual machine through the communication interface 23, the processor 22 is specifically configured to: process the UE processed by the second virtual machine by using the communication interface 23. The service data stream is sent to the first virtual machine.
  • the service switch of this embodiment may be used to implement the technical solutions of the foregoing method embodiments of the present invention, and the implementation principles and technical effects thereof are similar, and details are not described herein again.
  • the foregoing program may be stored in a computer readable storage medium, and the program is executed when executed.
  • the foregoing storage medium includes: read-only memory (English: Read-Only Memory, ROM for short), random access memory (English: Random Access Memory, RAM), disk or A variety of media such as optical discs that can store program code.

Landscapes

  • Data Exchanges In Wide-Area Networks (AREA)

Abstract

本发明实施例提供一种负荷分担的方法和业务交换机，该方法包括：业务交换机接收用户设备UE的业务数据流；根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；接收所述第一虚拟机处理后的所述UE的业务数据流；将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。从而降低了UE的业务数据流无法得到应用业务处理的风险，避免了所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。

Description

负荷分担的方法和业务交换机 技术领域
本发明实施例涉及通信技术领域，尤其涉及一种负荷分担的方法和业务交换机。
背景技术
业务链(英文：Service Function Chain，简称：SFC)是指基于用户、业务或者网络状态对业务进行智能分流，并按需将每个分支的流量自动转发给一个或多个增值业务的一种解决方案。图1为现有技术中的用户设备(User Equipment，简称：UE)访问网络的业务数据流的一种网络架构示意图，如图1所示，UE A和UE B访问网络(英文：Internet)的业务数据流经过分类器(英文：Classifier)，根据用户、业务和网络状态分流后都需要经过同一个业务链处理。图1所示的业务链中集成了两个应用业务，即应用业务1和应用业务2，其中，应用业务例如为：安全计费、安全防护、内容优化、内容过滤、缓存服务、防火墙等；UE A和UE B访问网络的业务数据流均要进行上述两个应用业务处理，即UE A和UE B访问网络的业务数据流先进行应用业务1的处理，然后进行应用业务2的处理；而且这种先进行应用业务1处理再进行应用业务2处理的处理序列，称之为逻辑链，可以表示为(应用业务1->应用业务2)。
如图1所示的处理应用业务1的实例为两个，处理应用业务2的实例(即虚拟机)为两个，分别为实例1和实例2，而数据由各个应用业务对应的哪个实例处理的处理序列，称之为物理链，实例是指用于提供应用业务的虚拟机。因此，确定由处理各个应用业务的哪个实例来对业务数据流进行应用业务处理的过程为：针对每个应用业务，均根据UE的IP地址来计算哈希值，然后根据哈希值确定处理应用业务的某个实例进行业务处理；对于同一个UE来说，针对每个应用业务计算出的哈希值均是相同的，例如：针对应用业务1，根据UE A的IP地址计算得到哈希值为1，此时选择处理应用业务1的实例1来对UE A的业务数据流进行应用业务1的处理；针对应用业务2，根据 UE A的IP地址计算得到哈希值为1，此时选择处理应用业务2的实例1来对UE B的业务数据流进行应用业务2的处理；因此UE A的物理链为(应用业务1：实例1->应用业务2：实例1)。针对应用业务1，根据UE B的IP地址计算得到哈希值为2，此时选择处理应用业务1的实例2来对UE B的业务数据流进行应用业务1的处理；针对应用业务2，根据UE B的IP地址计算得到哈希值为2，此时选择处理应用业务2的实例2来对UE B的业务数据流进行应用业务2的处理；因此UE B的物理链为(应用业务1：实例2->应用业务2：实例2)。
然而，现有技术中处理应用业务1的实例1与处理应用业务2的实例2对UE的业务数据流进行处理，或者是处理应用业务1的实例2与处理应用业务2的实例2对UE的业务数据流进行处理；若上述任一实例发生故障，将导致UE的业务数据流有一半的概率无法得到处理；而且若处理应用业务1的实例1和处理应用业务2的实例2同时发生故障，将导致所有UE的业务数据流均无法得到处理，造成所有的应用业务处理均发生中断；因此，降低了整个网络系统的可靠性。
发明内容
本发明实施例提供一种负荷分担的方法和业务交换机，用于降低UE的业务数据流无法得到应用业务处理的风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
第一方面，本发明实施例提供一种负荷分担的方法，包括：业务交换机接收UE的业务数据流；根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；然后将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；并接收所述第一虚拟机处理后的所述UE的业务数据流；再将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。由于本实施例中用于处理UE的业务数据流的虚拟机是通过UE的标识信息，以及与该虚拟机提供的应用业务的指示信息共同确定的，因此一个应用业务的虚拟机可能与其它应用业务的各个虚拟机分别属于处理业务数据流的不同物理链，在如此形成的物理链中若其中一个虚拟机 发生故障，而该物理链中的其它虚拟机可能也同时属于其它物理链，避免了其它虚拟机无法处理业务数据流的问题；而且只要各个应用业务中存在一个正常的虚拟机，这些正常的虚拟机会形成一个处理业务数据流的物理链，从而降低业务数据流无法得到应用业务处理的风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
根据第一方面，在第一方面的第一种可能的实现方式中，所述指示信息为所述第一应用业务的标识；或者，
所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
根据第一方面或第一方面的第一种可能的实现方式，在第一方面的第二种可能的实现方式中，所述UE的标识信息为所述UE的IP地址或者设备号。
根据第一方面或第一方面的第一种可能的实现方式或第一方面的第二种可能的实现方式，在第一方面的第三种可能的实现方式中，所述业务交换机根据UE的标识信息以及第一应用业务的指示信息，从提供第一应用业务的至少两个虚拟机中确定第一虚拟机，包括：
所述业务交换机对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；
所述业务交换机根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
根据第一方面或第一方面的上述任意一种可能的实现方式，在第一方面的第四种可能的实现方式中，所述目标设备为所述UE或者互联网服务器。
根据第一方面或第一方面的上述任意一种可能的实现方式，在第一方面的第五种可能的实现方式中，所述业务交换机将所述UE的业务数据流发送给所述第一虚拟机之前，还包括：
所述业务交换机根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；
所述业务交换机将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；
所述业务交换机接收所述第二虚拟机处理后的所述UE的业务数据流；
所述业务交换机将所述UE的业务数据流发送给所述第一虚拟机，包括：所述业务交换机将所述第二虚拟机处理后的所述UE的业务数据流发送给所 述第一虚拟机。
第二方面，本发明实施例提供一种业务交换机，包括：接收单元，用于接收UE的业务数据流；确定单元，用于根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；发送单元，用于将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；所述接收单元，还用于接收所述第一虚拟机处理后的所述UE的业务数据流；所述发送单元，还用于将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。由于本实施例中用于处理UE的业务数据流的虚拟机是通过UE的标识信息，以及与该虚拟机提供的应用业务的指示信息共同确定的，因此一个应用业务的虚拟机可能与其它应用业务的各个虚拟机分别属于处理业务数据流的不同物理链，在如此形成的物理链中若其中一个虚拟机发生故障，而该物理链中的其它虚拟机可能也同时属于其它物理链，避免了其它虚拟机无法处理业务数据流的问题；而且只要各个应用业务中存在一个正常的虚拟机，这些正常的虚拟机会形成一个处理业务数据流的物理链，从而降低业务数据流无法得到应用业务处理的风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
根据第二方面，在第二方面的第一种可能的实现方式中，所述指示信息为所述第一应用业务的标识；或者，
所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
根据第二方面或第二方面的第一种可能的实现方式，在第二方面的第二种可能的实现方式中，所述UE的标识信息为所述UE的IP地址或者设备号。
根据第二方面或第二方面的第一种可能的实现方式或第二方面的第二种可能的实现方式，在第二方面的第三种可能的实现方式中，所述确定单元具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
根据第二方面或第二方面的上述任意一种可能的实现方式，在第二方面的第四种可能的实现方式中，所述目标设备为所述UE或者互联网服务器。
根据第二方面或第二方面的上述任意一种可能的实现方式，在第二方面 的第五种可能的实现方式中，所述确定单元还用于在所述发送单元将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；
所述发送单元，还用于将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；
所述接收单元，还用于接收所述第二虚拟机处理后的所述UE的业务数据流；
所述发送单元在将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
第三方面，本发明实施例提供一种业务交换机，包括：存储器、处理器和通信接口；
所述存储器，用于存储执行负荷分担的方法的程序代码；
所述处理器，用于调用所述存储器中的所述程序代码执行：
通过所述通信接口接收用户设备UE的业务数据流；
根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；
通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；
通过所述通信接口接收所述第一虚拟机处理后的所述UE的业务数据流；
通过所述通信接口将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
根据第三方面，在第三方面的第一种可能的实现方式中，所述指示信息为所述第一应用业务的标识；或者，
所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
根据第三方面或第三方面的第一种可能的实现方式，在第三方面的第二种可能的实现方式中，所述UE的标识信息为所述UE的IP地址或者设备号。
根据第三方面或第三方面的第一种可能的实现方式或第三方面的第二种可能的实现方式，在第三方面的第三种可能的实现方式中，所述处理器在根 据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机时，具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
根据第三方面或第三方面的上述任意一种可能的实现方式，在第三方面的第四种可能的实现方式中，所述目标设备为所述UE或者互联网服务器。
根据第三方面或第三方面的上述任意一种可能的实现方式，在第三方面的第五种可能的实现方式中，所述处理器还用于，在通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；通过所述通信接口将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；通过所述通信接口接收所述第二虚拟机处理后的所述UE的业务数据流；
所述处理器在通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：通过所述通信接口将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
通过本发明实施例提供的上述方案，由于本实施例中用于处理UE的业务数据流的虚拟机是通过UE的标识信息，以及与该虚拟机提供的应用业务的指示信息共同确定的，因此一个应用业务的虚拟机可能与其它应用业务的各个虚拟机分别属于处理业务数据流的不同物理链，在如此形成的物理链中若其中一个虚拟机发生故障，而该物理链中的其它虚拟机可能也同时属于其它物理链，避免了其它虚拟机无法处理业务数据流的问题；而且只要各个应用业务中存在一个正常的虚拟机，这些正常的虚拟机会形成一个处理业务数据流的物理链，从而降低业务数据流无法得到应用业务处理的风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
附图说明
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一简单地介绍，显而易见地，下面描述中的附图是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1为现有技术中的UE访问网络的业务数据流的一种网络架构示意图；
图2为本发明实施例一提供的负荷分担的方法的流程图；
图3为本发明实施例二提供的负荷分担的方法的流程图；
图4为本发明实施例提供的形成物理链的一种示意图；
图5为本发明实施例一提供的业务交换机的结构示意图；
图6为本发明实施例二提供的业务交换机的结构示意图。
具体实施方式
为使本发明实施例的目的、技术方案和优点更加清楚，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，所描述的实施例是本发明一部分实施例，而不是全部的实施例。基于本发明中的实施例，本领域普通技术人员在没有作出创造性劳动前提下所获得的所有其他实施例，都属于本发明保护的范围。
本发明下述各实施例可以应用于图1所示的网络架构中，相应地，如图1所示的业务交换机1可以执行本发明各实施例提供的实现方案，如图1所示的业务交换机2也可以执行本发明各实施例提供的实现方案。
图2为本发明实施例一提供的负荷分担的方法的流程图，如图2所示，本实施例的方法可以包括：
S101、业务交换机接收UE的业务数据流。
S102、业务交换机根据UE的标识信息，以及第一应用业务的指示信息，从提供第一应用业务的至少两个虚拟机中确定第一虚拟机。
本实施例中，业务交换机接收到UE访问网络的业务数据流后，可以确定当前需要对UE访问网络的业务数据流进行第一应用业务的处理，此处的第一应用业务可以为需要处理该UE的业务数据流的任一应用业务。若提供第一应用业务的实例包括至少两个，实例是指用于提供应用业务的虚拟机，要实现对UE的业务数据流的第一应用业务的处理，需要从提供第一应用业务的至少两个虚拟机中确定出一个虚拟机，由确定出的该虚拟机对UE的业 务数据流进行第一应用业务的处理。然后业务交换机根据UE的标识信息，以及第一应用业务的指示信息，从提供第一应用业务的至少两个虚拟机中确定用于对该UE的业务数据流进行第一应用业务的处理的虚拟机，该虚拟机称为第一虚拟机；第一虚拟机为用于处理UE的业务数据流的物理链上的一个虚拟机，物理链是指依次处理该UE的业务数据流的各个应用业务的虚拟机的序列。
可选地，业务交换机接收的UE的业务数据流可以是该业务交换机的在该业务数据流的流向上的上一个设备发送的，或者是提供第二应用业务的一个虚拟机发送的，该第二应用业务为对UE的业务数据流进行应用业务处理的序列中第一应用业务的上一个应用业务。
可选地，UE的标识信息为该UE的IP地址或者设备号，或者其它用于唯一标识该UE的信息。
S103、业务交换机将UE的业务数据流发送给第一虚拟机，以使第一虚拟机对UE的业务数据流进行第一应用业务的处理。
S104、业务交换机接收第一虚拟机处理后的UE的业务数据流。
S105、业务交换机将第一虚拟机处理后的UE的业务数据流发送给目标设备。
本实施例中，业务交换机在确定第一虚拟机后，将接收的该UE的业务数据流发送给该第一虚拟机，由该第一虚拟机对该UE的业务数据流进行第一应用业务的处理，该第一虚拟机再将该第一虚拟机处理后的该UE的业务数据流发送给业务交换机，业务交换机接收到第一虚拟机处理后的该UE的业务数据流之后，将该第一虚拟机处理后的该UE的业务数据流发送给目标设备。
如果UE的业务数据流是从UE发往互联网的方向，则上述的目标设备为互联网服务器，即业务交换机将经过应用业务的处理后的UE的业务数据流发送给互联网服务器。如果UE的业务数据流是从互联网发往UE的方向，则上述的目标设备为UE，即业务交换机将经过应用业务的处理后的UE的业务数据流发送给UE。
可选地，S102的一种具体实现方式为：业务交换机对该UE的标识信息和该第一应用业务的指示信息进行哈希运算，获得哈希值，然后根据该哈希 值，从提供第一应用业务的至少两个虚拟机中确定与该哈希值对应的第一虚拟机。针对同一UE来说，由于对于不同的应用业务，其指示信息不同，因此运算获得的哈希值可能相同也可能不相同；从而一个应用业务中的一个哈希值对应的虚拟机与其它应用业务中的不同的哈希值对应的虚拟机共同对该UE的业务数据流进行应用业务处理，也可与其它应用业务中的相同的哈希值对应的虚拟机共同对该UE的业务数据流进行应用业务处理。
本实施例提供的负荷分担的方法，由于本实施例中用于处理UE的业务数据流的虚拟机是通过UE的标识信息，以及与该虚拟机提供的应用业务的指示信息共同确定的，因此一个应用业务的虚拟机可能与其它应用业务的各个虚拟机分别属于处理业务数据流的不同物理链，在如此形成的物理链中若其中一个虚拟机发生故障，而该物理链中的其它虚拟机可能也同时属于其它物理链，避免了其它虚拟机无法处理业务数据流的问题；而且只要各个应用业务中存在一个正常的虚拟机，这些正常的虚拟机会形成一个处理业务数据流的物理链，从而降低业务数据流无法得到应用业务处理的风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
图3为本发明实施例二提供的负荷分担的方法的流程图，如图3所示，本实施例的方法可以包括：
S201、业务交换机接收UE的第一业务数据流。
S202、业务交换机根据UE的标识信息以及第二应用业务的指示信息，从提供第二应用业务的处理的至少两个虚拟机中确定第二虚拟机。
S203、业务交换机将UE的第一业务数据流发送给第二虚拟机。
本实施例中，业务交换机可以确定需要对UE的业务数据流先进行第二应用业务的处理再进行第一应用业务的处理。业务交换机接收UE的第一业务数据流，然后根据UE的标识信息以及第二应用业务的指示信息，从提供第二应用业务的处理的至少两个虚拟机中确定一个虚拟机，该虚拟机为第二虚拟机。然后，业务交换机将该UE的第一业务数据流发送给第二虚拟机，以使第二虚拟机对该UE的第一业务数据流进行第二应用业务的处理。第二虚拟机接收到该UE的第一业务数据流后，对该UE的第一业务数据流进行第二应用业务的处理，获得该UE的第二业务数据流，该UE的第二业务数据流为该第二虚拟机处理后的该UE的第一业务数据流；第二虚拟机再将该第二 业务数据流发送给业务交换机。
可选地，第一业务数据流可以该业务交换机的上一跳设备发送给该业务交换机的，也可以是第三虚拟机将该第三虚拟机处理后的该UE的业务数据流发送给该业务交换机的，第三虚拟机为提供第三应用业务的虚拟机，该第三应用业务为对UE的业务数据流进行应用业务处理的序列中第二应用业务的上一个应用业务。
需要说明的是，业务交换机根据UE的标识信息以及第二应用业务的指示信息从提供第二应用业务的处理的至少两个虚拟机中确定第二虚拟机的实现方案，与，业务交换机根据UE的标识信息以及第一应用业务的指示信息从提供第一应用业务的处理的至少两个虚拟机中确定第一虚拟机的实现方案类似，此处不再赘述。
S204、业务交换机接收第二虚拟机发送的UE的第二业务数据流。
本实施例中，业务交换机接收第二虚拟机发送的该UE的第二业务数据流。
S205、业务交换机根据UE的标识信息以及第一应用业务的指示信息，从提供第一应用业务的至少两个虚拟机中确定第一虚拟机。
本实施例中，S205的实现过程可以参见本发明方法实施例一中的相关描述，此处不再赘述。需要说明的是，S205在S206之前执行即可，本实施例不限定S205在S204之后执行。
S206、业务交换机将UE的第二业务数据流发送给第一虚拟机。
S207、业务交换机接收第一虚拟机发送的UE的第三业务数据流。
本实施例中，业务交换机在确定第一虚拟机后，将第二业务数据流(即第二虚拟机处理后的第一业务数据流)发送给第一虚拟机，第一虚拟机接收到第二业务数据流后，对第二业务数据流进行第一应用业务的处理，获得第三业务数据流，该第三业务数据流为第一虚拟机处理后的第二业务数据流，也就是第二虚拟机处理后和第一虚拟机处理后的第一业务数据流。
S208、业务交换机将UE的第三业务数据流发送给目标设备。
本实施例中，S208的实现过程可以参见本发明方法实施例一中的相关描述，此处不再赘述。
本实施例通过上述方案，可以降低业务数据流无法得到应用业务处理的 风险，避免所有的应用业务处理均发生中断的现象，提高整个网络系统的可靠性。
下面以具体的举子对本发明的方案中确定提供应用业务的虚拟机进行详细介绍。
例如：UE1、UE2、UE3、UE4均经过同一业务链，并且对这四个UE的业务数据流进行应用业务的处理顺序为：先应用业务1再应用业务2；其中，提供应用业务1的虚拟机包括：应用业务1的虚拟机1和应用业务1的虚拟机2，提供应用业务2的虚拟机包括：应用业务1的虚拟机1和应用业务2的虚拟机2，若对各个UE的业务数据流进行应用业务1的处理和应用业务2的处理均由同一交换机转发，此处以应用业务的指示信息为应用业务的标识为例。
UE1的业务数据流到达业务交换机后，业务交换机根据UE1的IP地址和应用业务1的标识，确定应用业务1的虚拟机1对UE1的业务数据流进行应用业务1的处理；业务交换机还根据UE1的IP地址和应用业务2的标识，确定应用业务2的虚拟机1对UE1的业务数据流进行应用业务2的处理；因此，对UE1的业务数据流进行各个应用业务的处理的各个虚拟机的顺序为：先应用业务1的虚拟机1再应用业务2的虚拟机1。例如：如图4所示，业务交换机将UE1的IP地址为10.1.1.1以及应用业务1的标识为1进行哈希运算，获得的哈希值为1，哈希值为1对应的虚拟机为应用业务1的虚拟机1，因此业务交换机从提供第一应用业务的虚拟机中确定应用业务1的虚拟机1；业务交换机将UE1的IP地址为10.1.1.1以及应用业务2的标识为2进行哈希运算，获得的哈希值为1，哈希值为1对应的虚拟机为应用业务1的虚拟机1；因此，UE1的业务数据流先由应用业务1的虚拟机1处理再由应用业务2的虚拟机1处理的序列称为如图4所示的物理链1。
UE2的业务数据流到达业务交换机后，业务交换机根据UE2的IP地址和应用业务1的标识，确定应用业务1的虚拟机1对UE2的业务数据流进行应用业务1的处理；业务交换机还根据UE2的IP地址和应用业务2的标识，确定应用业务2的虚拟机2对UE2的业务数据流进行应用业务2的处理；因此，对UE2的业务数据流进行各个应用业务的处理的各个虚拟机的顺序为：先应用业务1的虚拟机1再应用业务2的虚拟机2。例如：如图4所示，业 务交换机将UE2的IP地址为10.1.1.2以及应用业务1的标识为1进行哈希运算，获得的哈希值为1，哈希值为1对应的虚拟机为第一应用业务的虚拟机1，因此业务交换机从提供应用业务1的虚拟机中确定应用业务1的虚拟机1；业务交换机将UE2的IP地址为10.1.1.2以及应用业务2的标识为2进行哈希运算，获得的哈希值为2，哈希值为2对应的虚拟机为应用业务2的虚拟机2；因此，UE2的业务数据流先由应用业务1的虚拟机1处理再由应用业务2的虚拟机2处理的序列称为如图4所示的物理链2。
UE3的业务数据流到达业务交换机后，业务交换机根据UE3的IP地址和应用业务1的标识，确定应用业务1的虚拟机2对UE3的业务数据流进行应用业务1的处理；业务交换机还根据UE3的IP地址和应用业务2的标识，确定应用业务2的虚拟机2对UE3的业务数据流进行应用业务2的处理；因此，对UE3的业务数据流进行各个应用业务的处理的各个虚拟机的顺序为：先应用业务1的虚拟机2再应用业务2的虚拟机2。例如：如图4所示，业务交换机将UE3的IP地址为10.1.1.3以及应用业务1的标识为1进行哈希运算，获得的哈希值为2，哈希值为2对应的虚拟机为应用业务1的虚拟机2，因此业务交换机从提供应用业务1的虚拟机中确定应用业务1的虚拟机2；业务交换机将UE3的IP地址为10.1.1.3以及应用业务2的标识为2进行哈希运算，获得的哈希值为2，哈希值为2对应的虚拟机为应用业务2的虚拟机2；因此，UE3的业务数据流先由应用业务1的虚拟机2处理再由应用业务2的虚拟机2处理的序列称为如图4所示的物理链3。
UE4的业务数据流到达业务交换机后，业务交换机根据UE4的IP地址和应用业务1的标识，确定应用业务1的虚拟机2对UE4的业务数据流进行应用业务1的处理；业务交换机还根据UE4的IP地址和应用业务2的标识，确定应用业务2的虚拟机1对UE4的业务数据流进行应用业务2的处理；因此，对UE4的业务数据流进行各个应用业务的处理的各个虚拟机的顺序为：先应用业务1的虚拟机2再应用业务2的虚拟机1。例如：如图4所示，业务交换机将UE4的IP地址为10.1.1.4以及应用业务1的标识为1进行哈希运算，获得的哈希值为2，哈希值为2对应的虚拟机为应用业务1的虚拟机2，因此业务交换机从提供应用业务1的虚拟机中确定应用业务1的虚拟机2；业务交换机将UE4的IP地址为10.1.1.4以及应用业务2的标识为2进行哈希 运算，获得的哈希值为1，哈希值为1对应的虚拟机为应用业务2的虚拟机1；因此，UE4的业务数据流先由应用业务1的虚拟机2处理再由应用业务2的虚拟机1处理的序列称为如图4所示的物理链4。
需要说明的是，确定对各个UE的业务数流流进行应用业务1的处理的虚拟机的业务交换机与对各个UE的业务数据流进行应用业务2的处理的虚拟机的业务交换机可以不是同一台业务交换机。
通过上述举例可知，业务交换机根据UE的IP地址和应用业务的标识确定进行应用业务的处理的虚拟机，由此形成的处理UE的业务数据流的物理链为多条，因此若图4所示的应用业务1的虚拟机1出现了故障，由于应用业务2的虚拟机1和应用业务2的虚拟机2分别与应用业务1的虚拟机2形成了处理业务数据流的物理链，因此不会影响到应用业务2的各虚拟机对业务数据流进行应用业务2的处理。
而在现有技术的方案中，由于业务交换机是根据UE的IP地址来确定提供应用业务的虚拟机，对同一个UE来说，要么是确定应用业务1的虚拟机1和应用业务2的虚拟机1，要么是确定应用业务1的虚拟机2和应用业务2的虚拟机2；一共可形成两条处理UE的业务数据流的物理链：一条物理链为：先应用业务1的虚拟机1再应用业务2的虚拟机1，另一条物理链为：先应用业务1的虚拟机2再应用业务2的虚拟机2；如果应用业务1的虚拟机1和应用业务2的虚拟机2出现了故障，则这两条物理链均不可用，从而导致所有的应用业务处理中断，影响了整个网络系统的可靠性。
但是通过本发明实施例提供的上述方案，即使应用业务1的虚拟机1和应用业务2的虚拟机2出现了故障，虽然影响了上述物理链1、2、3，但是上述物理链4仍然是正常的，降低了UE的业务数据流无法得到应用业务处理的风险，从而避免了所有的应用业务处理中断的情况，保证了整个网络系统的可靠性。
另外，若各个业务交换机上集成一个应用业务，则上述应用业务的指示信息可以为该业务交换机的标识或者该业务交换机的随机值，该业务交换机的标识用于区别于其它业务交换机，具体实现过程可以参见业务交换机根据上述应用业务的标识来确定应用业务的虚拟机中的相关记载，此处不再赘述。业务交换机的随机值是该业务交换机部署在网络中随机生成的一个值，这个 值用于区别于其它业务交换机。
图5为本发明实施例一提供的业务交换机的结构示意图，如图5所示，本实施例的业务交换机可以包括：接收单元11、确定单元12和发送单元13。其中，接收单元11，用于接收UE的业务数据流；确定单元12，用于根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；发送单元13，用于将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；接收单元11，还用于接收所述第一虚拟机处理后的所述UE的业务数据流；发送单元13，还用于将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
可选地，所述指示信息为所述第一应用业务的标识；或者，
所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
可选地，所述UE的标识信息为所述UE的IP地址或者设备号。
可选地，确定单元12具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
可选地，所述目标设备为所述UE或者互联网服务器。
可选地，确定单元12还用于在发送单元13将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；
发送单元13，还用于将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；
接收单元11，还用于接收所述第二虚拟机处理后的所述UE的业务数据流；
发送单元13在将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
本实施例的业务交换机，可以用于执行本发明上述各方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
图6为本发明实施例二提供的业务交换机的结构示意图，如图6所示，本实施例的业务交换机可以包括：存储器21、处理器22和通信接口23；其中，存储器21，用于存储执行负荷分担的方法的程序代码；处理器22，用于调用存储器21中的所述程序代码执行：
通过通信接口23接收UE的业务数据流；
根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；
通过通信接口23将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；
通过通信接口23接收所述第一虚拟机处理后的所述UE的业务数据流；
通过通信接口23将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
可选地，所述指示信息为所述第一应用业务的标识；或者，
所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
可选地，所述UE的标识信息为所述UE的IP地址或者设备号。
可选地，处理器22在根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机时，具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
可选地，所述目标设备为所述UE或者互联网服务器。
可选地，处理器22还用于，在通过通信接口23将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；通过通信接口23将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；以及通过通信接口23接收所述第二虚拟机处理后的所述UE的业务数据流；
处理器22在通过通信接口23将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：通过通信接口23将所述第二虚拟机处理后的所述UE 的业务数据流发送给所述第一虚拟机。
本实施例的业务交换机，可以用于执行本发明上述各方法实施例的技术方案，其实现原理和技术效果类似，此处不再赘述。
本领域普通技术人员可以理解：实现上述方法实施例的全部或部分步骤可以通过程序指令相关的硬件来完成，前述的程序可以存储于一计算机可读取存储介质中，该程序在执行时，执行包括上述方法实施例的步骤；而前述的存储介质包括：只读内存(英文：Read-Only Memory，简称：ROM)、随机存取存储器(英文：Random Access Memory，简称：RAM)、磁碟或者光盘等各种可以存储程序代码的介质。
最后应说明的是：以上各实施例仅用以说明本发明的技术方案，而非对其限制；尽管参照前述各实施例对本发明进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本发明各实施例技术方案的范围。

Claims (18)

  1. 一种负荷分担的方法，其特征在于，包括：
    业务交换机接收用户设备UE的业务数据流；
    所述业务交换机根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；
    所述业务交换机将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；
    所述业务交换机接收所述第一虚拟机处理后的所述UE的业务数据流；
    所述业务交换机将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
  2. 根据权利要求1所述的方法，其特征在于，所述指示信息为所述第一应用业务的标识；或者，
    所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
  3. 根据权利要求1或2所述的方法，其特征在于，所述UE的标识信息为所述UE的互联网协议IP地址或者设备号。
  4. 根据权利要求1-3任意一项所述的方法，其特征在于，所述业务交换机根据UE的标识信息以及第一应用业务的指示信息，从提供第一应用业务的至少两个虚拟机中确定第一虚拟机，包括：
    所述业务交换机对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；
    所述业务交换机根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
  5. 根据权利要求1-4任意一项所述的方法，其特征在于，所述目标设备为所述UE或者互联网服务器。
  6. 根据权利要求1-5任意一项所述的方法，其特征在于，所述业务交换机将所述UE的业务数据流发送给所述第一虚拟机之前，还包括：
    所述业务交换机根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；
    所述业务交换机将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；
    所述业务交换机接收所述第二虚拟机处理后的所述UE的业务数据流；
    所述业务交换机将所述UE的业务数据流发送给所述第一虚拟机，包括：所述业务交换机将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
  7. 一种业务交换机，其特征在于，包括：
    接收单元，用于接收用户设备UE的业务数据流；
    确定单元，用于根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；
    发送单元，用于将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；
    所述接收单元，还用于接收所述第一虚拟机处理后的所述UE的业务数据流；
    所述发送单元，还用于将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
  8. 根据权利要求7所述的业务交换机，其特征在于，所述指示信息为所述第一应用业务的标识；或者，
    所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
  9. 根据权利要求7或8所述的业务交换机，其特征在于，所述UE的标识信息为所述UE的互联网协议IP地址或者设备号。
  10. 根据权利要求7-9任意一项所述的业务交换机，其特征在于，所述确定单元具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
  11. 根据权利要求7-10任意一项所述的业务交换机，其特征在于，所述目标设备为所述UE或者互联网服务器。
  12. 根据权利要求7-11任意一项所述的业务交换机，其特征在于，所述确定单元还用于在所述发送单元将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；
    所述发送单元，还用于将所述UE的业务数据流发送给所述第二虚拟机， 以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；
    所述接收单元，还用于接收所述第二虚拟机处理后的所述UE的业务数据流；
    所述发送单元在将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
  13. 一种业务交换机，其特征在于，包括：存储器、处理器和通信接口；
    所述存储器，用于存储执行负荷分担的方法的程序代码；
    所述处理器，用于调用所述存储器中的所述程序代码执行：
    通过所述通信接口接收用户设备UE的业务数据流；
    根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机；
    通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机，以使所述第一虚拟机对所述UE的业务数据流进行所述第一应用业务的处理；
    通过所述通信接口接收所述第一虚拟机处理后的所述UE的业务数据流；
    通过所述通信接口将所述第一虚拟机处理后的所述UE的业务数据流发送给目标设备。
  14. 根据权利要求13所述的业务交换机，其特征在于，所述指示信息为所述第一应用业务的标识；或者，
    所述指示信息为所述业务交换机的标识或者所述业务交换机的随机值。
  15. 根据权利要求13或14所述的业务交换机，其特征在于，所述UE的标识信息为所述UE的互联网协议IP地址或者设备号。
  16. 根据权利要求13-15任意一项所述的业务交换机，其特征在于，所述处理器在根据所述UE的标识信息以及第一应用业务的指示信息，从提供所述第一应用业务的至少两个虚拟机中确定第一虚拟机时，具体用于：对所述UE的标识信息和所述指示信息进行哈希运算，获得哈希值；以及根据所述哈希值，从提供所述第一应用业务的至少两个虚拟机中确定与所述哈希值对应的所述第一虚拟机。
  17. 根据权利要求13-16任意一项所述的业务交换机，其特征在于，所述目标设备为所述UE或者互联网服务器。
  18. 根据权利要求13-17任意一项所述的业务交换机，其特征在于，所述处理器还用于，在通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机之前，根据所述UE的标识信息以及第二应用业务的指示信息，从提供所述第二应用业务的处理的至少两个虚拟机中确定所述第二虚拟机；通过所述通信接口将所述UE的业务数据流发送给所述第二虚拟机，以使所述第二虚拟机对所述UE的业务数据流进行所述第二应用业务的处理；通过所述通信接口接收所述第二虚拟机处理后的所述UE的业务数据流；
    所述处理器在通过所述通信接口将所述UE的业务数据流发送给所述第一虚拟机时，具体用于：通过所述通信接口将所述第二虚拟机处理后的所述UE的业务数据流发送给所述第一虚拟机。
PCT/CN2015/100237 2015-12-31 2015-12-31 负荷分担的方法和业务交换机 WO2017113346A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/100237 WO2017113346A1 (zh) 2015-12-31 2015-12-31 负荷分担的方法和业务交换机

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2015/100237 WO2017113346A1 (zh) 2015-12-31 2015-12-31 负荷分担的方法和业务交换机

Publications (1)

Publication Number Publication Date
WO2017113346A1 true WO2017113346A1 (zh) 2017-07-06

Family

ID=59224295

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2015/100237 WO2017113346A1 (zh) 2015-12-31 2015-12-31 负荷分担的方法和业务交换机

Country Status (1)

Country Link
WO (1) WO2017113346A1 (zh)

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110055845A1 (en) * 2009-08-31 2011-03-03 Thyagarajan Nandagopal Technique for balancing loads in server clusters
CN103929492A (zh) * 2014-04-28 2014-07-16 华为技术有限公司 业务链负载均衡方法及其装置、系统
CN103929368A (zh) * 2014-05-05 2014-07-16 华为技术有限公司 多业务单元负载均衡方法及装置
CN104283806A (zh) * 2014-10-31 2015-01-14 杭州华三通信技术有限公司 业务链处理方法和设备
CN104954274A (zh) * 2014-03-25 2015-09-30 华为技术有限公司 生成转发信息的方法、控制器和业务转发实体
CN105207909A (zh) * 2014-06-24 2015-12-30 华为技术有限公司 一种发送信息的方法和网络装置

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110055845A1 (en) * 2009-08-31 2011-03-03 Thyagarajan Nandagopal Technique for balancing loads in server clusters
CN104954274A (zh) * 2014-03-25 2015-09-30 华为技术有限公司 生成转发信息的方法、控制器和业务转发实体
CN103929492A (zh) * 2014-04-28 2014-07-16 华为技术有限公司 业务链负载均衡方法及其装置、系统
CN103929368A (zh) * 2014-05-05 2014-07-16 华为技术有限公司 多业务单元负载均衡方法及装置
CN105207909A (zh) * 2014-06-24 2015-12-30 华为技术有限公司 一种发送信息的方法和网络装置
CN104283806A (zh) * 2014-10-31 2015-01-14 杭州华三通信技术有限公司 业务链处理方法和设备

Similar Documents

Publication Publication Date Title
US20190036819A1 (en) Use of hypervisor for active-active stateful network service cluster
US10191758B2 (en) Directing data traffic between intra-server virtual machines
US9083645B2 (en) Systems and methods providing reverse path forwarding compliance for a multihoming virtual routing bridge
US9930013B2 (en) Control of out-of-band multipath connections
EP2676411B1 (en) Method and system for classification and management of inter-blade network traffic in a blade server
US20160119253A1 (en) Method and system of performing service function chaining
US9614739B2 (en) Defining service chains in terms of service functions
EP3069484B1 (en) Shortening of service paths in service chains in a communications network
US10771475B2 (en) Techniques for exchanging control and configuration information in a network visibility system
US9385950B2 (en) Configurable service proxy local identifier mapping
CN105745883B (zh) 转发表同步的方法、网络设备和系统
WO2016029749A1 (zh) 一种通信故障的检测方法、装置及系统
EP3399723B1 (en) Performing upper layer inspection of a flow based on a sampling rate
WO2015106729A1 (zh) 一种负载均衡方法、装置和系统、计算机存储介质
US10951584B2 (en) Methods for active-active stateful network service cluster
US11616724B2 (en) Packet transmission method and apparatus
JP2016528630A (ja) アプリケーションアウェアネットワーク管理
WO2016107379A1 (zh) 一种发送报文的方法和装置
US11968080B2 (en) Synchronizing communication channel state information for high flow availability
JP2008293492A (ja) ロードバランス型ネットワーク環境におけるインテリジェントフェイルバック
CN105991347A (zh) Dns请求报文的重定向方法和装置
WO2016197689A1 (zh) 处理报文的方法、装置和系统
WO2020046728A1 (en) Path management for segment routing based mobile user-plane using seamless bfd
WO2017032223A1 (zh) 虚拟机部署方法及装置
US10581669B2 (en) Restoring control-plane connectivity with a network management entity

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: 15911972

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 15911972

Country of ref document: EP

Kind code of ref document: A1