(19)
(11)EP 3 361 684 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
29.07.2020 Bulletin 2020/31

(21)Application number: 15907031.7

(22)Date of filing:  31.10.2015
(51)International Patent Classification (IPC): 
H04L 12/803(2013.01)
H04L 12/26(2006.01)
H04L 12/751(2013.01)
H04L 12/725(2013.01)
H04L 12/841(2013.01)
H04L 12/851(2013.01)
(86)International application number:
PCT/CN2015/093525
(87)International publication number:
WO 2017/070970 (04.05.2017 Gazette  2017/18)

(54)

ROUTE DETERMINING METHOD AND CORRESPONDING APPARATUS AND SYSTEM

ROUTENBESTIMMUNGSVERFAHREN UND ZUGEHÖRIGE VORRICHTUNG UND SYSTEM

PROCÉDÉ DE DÉTERMINATION DE ROUTE ET APPAREIL ET SYSTÈME CORRESPONDANTS


(84)Designated Contracting States:
AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TR

(43)Date of publication of application:
15.08.2018 Bulletin 2018/33

(73)Proprietor: Huawei Technologies Co., Ltd.
Longgang District Shenzhen, Guangdong 518129 (CN)

(72)Inventors:
  • TAN, Shiyong
    Shenzhen Guangdong 518129 (CN)
  • QIAO, Xiaoqiang
    Shenzhen Guangdong 518129 (CN)

(74)Representative: Thun, Clemens 
Mitscherlich PartmbB Patent- und Rechtsanwälte Sonnenstraße 33
80331 München
80331 München (DE)


(56)References cited: : 
EP-A1- 1 715 715
CN-A- 104 158 755
CN-A- 104 168 209
CN-A- 103 929 492
CN-A- 104 158 755
  
      
    Note: Within nine months from the publication of the mention of the grant of the European patent, any person may give notice to the European Patent Office of opposition to the European patent granted. Notice of opposition shall be filed in a written reasoned statement. It shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention).


    Description

    TECHNICAL FIELD



    [0001] The present invention relates to the field of communications technologies, and in particular, to a route determining method, and a corresponding apparatus and system.

    BACKGROUND



    [0002] In the field of communications technologies, a service flow initiated by a user usually needs to be processed by using multiple types of service functions (English: Service Function, SF for short), and different types of service flows may require different service functions.

    [0003] In the prior art, different types of service flows are classified into different service chains. SF sequences that each service chain needs to pass through and an order of passing through the SF sequences are defined. Service flows of different service chains need to pass through only corresponding service nodes. During specific implementation, a classifier classifies the service flows initiated by the user, and adds a service chain identifier to a packet of the service flow. Then the classifier sends the packet to a forwarder, and the forwarder determines, according to the service chain identifier included in the packet and a route mapping relationship of the service flow of the service chain, a next service node for receiving the packet.

    [0004] Actually, a same type of SFs may include multiple service nodes. When the forwarder forwards a packet, load balancing between multiple service nodes of a same type needs to be considered. In addition, dynamic load of the multiple service nodes varies with time. However, the prior art does not provide an efficient and flexible method for implementing service node load balancing.

    [0005] CN103929492A relates to load balancing wherein service chain selection and control policy is used to instruct a flow classifier to hash a service flow passing through the service chain, to obtain multiple subflows, wherein service chain identifiers and hashing factors are added to packets of the subflows. Thus, the different subflows correspond to and include different hashing factors, and by sending the packets of subflows to a forwarding device, the forwarding device can use the hashing factors to allocate each subflow to a value-added service device participating in a service of the same type.

    [0006] CN104158755A relates to a routing method wherein a dispatching device receives a target packet sent by user equipment. The target packet carries time indication information, and the time indication information is used to indicate an interval between a sending time of the target packet and a sending time of the first packet of the underlying target service. This time information is used for determining a target service node from at least two service nodes according to the receiving time of the first packet of the target service, and the target packet is forwarded accordingly.

    SUMMARY



    [0007] The present invention is defined in the independent claims. The dependent claims define particular embodiments of the invention.

    [0008] Embodiments of this application provide a route determining method, and a corresponding apparatus and system, so as to resolve a prior-art problem that it is difficult to implement service node load balancing efficiently and flexibly.

    [0009] According to a first aspect, an embodiment of this application provides a route determining method, including:

    receiving a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry;

    receiving a second flow entry, where the second flow entry includes a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period;

    receiving a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    determining a time period that includes the first timestamp from the first effective time period and the second effective time period;

    determining, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and

    sending the first packet to the determined service node.



    [0010] In the foregoing implementation, a switch receives the first flow entry and the second flow entry, and determines, according to the timestamp included in the first packet, whether to use the first flow entry or the second flow entry to determine a route for the first packet. Therefore, when a new flow entry is generated because a load status of a service node changes, the switch can immediately use the new flow entry to forward a packet of a newly created service flow, and continue to forward a packet of a previously created service flow based on a previous flow entry. Different from the prior art in which a packet is forwarded according to a new flow entry only after an existing packet is processed, in the solution provided in this embodiment of this application, the switch can adjust a routing policy in a timely manner and implement load balancing more flexibly and quickly. In addition, in the solution provided in this embodiment of this application, a flow entry is generated for a same type of service flows instead of each service flow. Therefore, there are few flow entries, system overheads are relatively low, and efficiency is relatively high.
    With reference to the first aspect, in a first possible implementation of the first aspect,
    after the receiving a second flow entry, the method further includes:

    determining whether a start time of the second effective time period precedes an end time of the first effective time period; and

    if the start time of the second effective time period precedes the end time of the first effective time period, changing the end time of the first effective time period to the start time of the second effective time period.



    [0011] In the foregoing possible implementation, when the subsequently generated second flow entry is in conflict with the previously generated first flow entry, an effective time of the first flow entry is changed according to an effective time of the second flow entry, to avoid a conflict between the flow entries.

    [0012] With reference to the first aspect or the first possible implementation of the first aspect, in a second possible implementation of the first aspect, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    the determining, according to the first flow entry, a service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generating a load balancing factor according to the first timestamp; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0013] With reference to the first aspect or the first possible implementation of the first aspect, in a third possible implementation of the first aspect, the first packet further includes a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    the determining, according to the first flow entry, a service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0014] With reference to the second or the third possible implementation of the first aspect, in a fourth possible implementation of the first aspect, the first packet is a head packet of the first service flow, and after the determining, from the at least two service nodes, a first service node for receiving the first packet, the method further includes:

    receiving a second packet of the first service flow; and

    sending the second packet to the first service node.



    [0015] In the foregoing possible implementation, a packet subsequent to the head packet of the first service flow is directly forwarded in a routing manner of the head packet of the first service flow, without a need to determine routes for packets of the service flow one by one by means of calculation, so that operation load on the switch is greatly reduced.

    [0016] With reference to the fourth possible implementation of the first aspect, in a fifth possible implementation of the first aspect, after the determining, from the at least two service nodes, a first service node for receiving the first packet, the method further includes:

    starting a timeout timer; and

    before the sending the second packet to the first service node, the method further includes:
    determining that the timeout timer does not expire.



    [0017] According to a second aspect, an embodiment of this application provides a route determining method, including:

    receiving a first packet of a first service flow;

    determining, according to a feature field included in the first packet, that the first service flow is a service flow of a first service chain;

    adding a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and

    sending the first packet to a switch.



    [0018] In the foregoing possible implementation, when classifying and identifying packets of service flows, a classifier adds a timestamp to the packet of the service flow, to add a time attribute to the packet, so that the switch can use different routing policies for packets in different time periods. Therefore, service node load balancing is implemented flexibly and efficiently.

    [0019] With reference to the second aspect, in a first possible implementation of the second aspect, before the sending the first packet to a switch, the method further includes:
    determining a load balancing factor of the first packet, and adding the load balancing factor to the first packet.

    [0020] According to a third aspect, an embodiment of this application provides a route determining method, including:

    obtaining first current load of each service node existing at a first moment;

    generating a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node;

    sending the first flow entry to a switch;

    obtaining second current load of each service node existing at a second moment;

    generating a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period; and

    sending the second flow entry to the switch.



    [0021] In the foregoing possible implementation, when detecting a load change of a service node, a controller can immediately generate the second flow entry, and send the second flow entry to the switch, so that the switch can process an existing packet according to the previous first flow entry, and process a newly created packet according to the second flow entry, thereby implementing load balancing flexibly and efficiently.

    [0022] According to a fourth aspect, an embodiment of this application provides a route determining apparatus, including:

    a first receiving module, configured to: receive a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry; and receive a second flow entry, where the second flow entry includes a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period;

    a second receiving module, configured to receive a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    a first determining module, configured to determine a time period that includes the first timestamp from the first effective time period and the second effective time period;

    a second determining module, configured to determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and

    a sending module, configured to send the first packet to the determined service node.



    [0023] With reference to the fourth aspect, in a first possible implementation of the fourth aspect, the apparatus further includes:
    a change module, configured to: after the second flow entry is received, determine whether a start time of the second effective time period precedes an end time of the first effective time period; and when the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.

    [0024] With reference to the fourth aspect or the first possible implementation of the fourth aspect, in a second possible implementation of the fourth aspect, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    the second determining module is specifically configured to:

    determine at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generate a load balancing factor according to the first timestamp; and

    determine, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0025] With reference to the fourth aspect or the first possible implementation of the fourth aspect, in a third possible implementation of the fourth aspect, the first packet further includes a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    the second determining module is specifically configured to:

    determine at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determine, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0026] With reference to the second or the third possible implementation of the fourth aspect, in a fourth possible implementation of the fourth aspect, the first packet is a head packet of the first service flow;
    the second receiving module is further configured to receive a second packet of the first service flow; and
    the sending module is further configured to send the second packet to the first service node.

    [0027] With reference to the fourth possible implementation of the fourth aspect, in a fifth possible implementation of the fourth aspect, the apparatus further includes:
    a third determining module, configured to: start a timeout timer after the first service node for receiving the first packet is determined from the at least two service nodes; and before the second packet is sent to the first service node, determine that the timeout timer does not expire.

    [0028] According to a fifth aspect, an embodiment of this application provides a route determining apparatus, including:

    a receiving module, configured to receive a first packet of a first service flow;

    a determining module, configured to determine, according to a feature field included in the packet, that the first service flow is a service flow of a first service chain;

    an identification module, configured to add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and

    a sending module, configured to send the first packet to a switch.



    [0029] With reference to the fifth aspect, in a first possible implementation of the fifth aspect, the identification module is further configured to: determine a load balancing factor of the first packet, and add the load balancing factor to the first packet.

    [0030] According to a sixth aspect, an embodiment of this application provides a route determining apparatus, including:

    a first obtaining module, configured to obtain first current load of each service node existing at a first moment;

    a first generation module, configured to generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node;

    a sending module, configured to send the first flow entry to a switch;

    a second obtaining module, configured to obtain second current load of each service node existing at a second moment; and

    a second generation module, configured to generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period, where

    the sending module is further configured to send the second flow entry to the switch.



    [0031] According to a seventh aspect, an embodiment of this application provides a route determining device, including:

    a receiver, configured to: receive a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry; receive a second flow entry, where the second flow entry includes a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period; and receive a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    a processor, configured to: determine a time period that includes the first timestamp from the first effective time period and the second effective time period; and determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and

    a transmitter, configured to send the first packet to the determined service node.



    [0032] With reference to the seventh aspect, in a first possible implementation of the seventh aspect, the processor is further configured to: after the second flow entry is received, determine whether a start time of the second effective time period precedes an end time of the first effective time period; and if the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.

    [0033] With reference to the seventh aspect or the first possible implementation of the seventh aspect, in a second possible implementation of the seventh aspect, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    that the processor is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generating a load balancing factor according to the first timestamp; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0034] With reference to the seventh aspect or the first possible implementation of the seventh aspect, in a third possible implementation of the seventh aspect, the first packet further includes a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period; and
    that the processor is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0035] With reference to the second or the third possible implementation of the seventh aspect, in a fourth possible implementation of the seventh aspect, the first packet is a head packet of the first service flow;
    the receiver is further configured to receive a second packet of the first service flow; and
    the transmitter is further configured to send the second packet to the first service node.

    [0036] With reference to the fourth possible implementation of the seventh aspect, in a fifth possible implementation of the seventh aspect, the processor is further configured to: start a timeout timer after the first service node for receiving the first packet is determined from the at least two service nodes; and before the transmitter sends the second packet to the first service node, determine that the timeout timer does not expire.

    [0037] According to an eighth aspect, an embodiment of this application provides a route determining device, including:

    a receiver, configured to receive a first packet of a first service flow;

    a processor, configured to: determine, according to a feature field included in the first packet, that the first service flow is a service flow of a first service chain; and add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and

    a transmitter, configured to send the first packet to a switch.



    [0038] With reference to the eighth aspect, in a first possible implementation of the eighth aspect, the processor is further configured to: determine a load balancing factor of the first packet, and add the load balancing factor to the first packet.

    [0039] According to a ninth aspect, an embodiment of this application provides a route determining device, including:

    a receiver, configured to obtain first current load of each service node existing at a first moment;

    a processor, configured to generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node; and

    a transmitter, configured to send the first flow entry to a switch, where

    the receiver is further configured to obtain second current load of each service node existing at a second moment;

    the processor is further configured to generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period; and

    the transmitter is further configured to send the second flow entry to the switch.



    [0040] According to a tenth aspect, an embodiment of this application provides a route determining system, including:

    a controller, configured to: obtain first current load of each service node existing at a first moment; generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node; send the first flow entry to a switch; obtain second current load of each service node existing at a second moment; generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second load proportion is generated according to the second current load of each service node; and send the second flow entry to the switch, where the second effective time period is different from the first effective time period;

    a classifier, configured to: receive a first packet of a first service flow; determine, according to a feature field included in the first packet, that the first service flow is a service flow of the first service chain; add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and send the first packet to the switch; and

    the switch, configured to: receive the first flow entry; receive the second flow entry; receive the first packet of the first service flow; determine a time period that includes the first timestamp from the first effective time period and the second effective time period; determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and send the first packet to the determined service node.



    [0041] With reference to the tenth aspect, in a first possible implementation of the tenth aspect, the switch is further configured to: after receiving the second flow entry,
    determine whether a start time of the second effective time period precedes an end time of the first effective time period; and
    if the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.

    [0042] With reference to the tenth aspect or the first possible implementation of the tenth aspect, the first packet is a head packet of the first service flow; and
    the switch is further configured to: receive a second packet of the first service flow; and send the second packet to the first service node.

    [0043] With reference to the tenth aspect, or the first or the second possible implementation of the tenth aspect, in a third possible implementation of the tenth aspect, that the switch is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generating a load balancing factor according to the first timestamp; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, the first service node for receiving the first packet.



    [0044] With reference to the tenth aspect, or the first or the second possible implementation of the tenth aspect, in a fourth possible implementation of the tenth aspect, the first packet further includes a load balancing factor, and that the switch is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determining, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, the first service node for receiving the first packet.



    [0045] With reference to the second possible implementation of the tenth aspect, in a fifth possible implementation of the tenth aspect, the switch is further configured to start a timeout timer after determining, from the at least two service nodes, the first service node for receiving the first packet; and
    the switch is further configured to: before sending the second packet to the first service node, determine that the timeout timer does not expire.

    BRIEF DESCRIPTION OF DRAWINGS



    [0046] To describe the technical solutions in the embodiments of this application more clearly, the following briefly describes the accompanying drawings required for describing the embodiments. Apparently, the accompanying drawings in the following description show merely some embodiments of the present invention, and persons of ordinary skill in the art may still derive other drawings from these accompanying drawings without creative efforts.

    FIG. 1 is a schematic diagram of a packet routing system;

    FIG. 2A and FIG. 2B are a schematic flowchart of a route determining method according to an embodiment of this application;

    FIG. 3 is a schematic flowchart of a further detailed route determining method according to an embodiment of this application;

    FIG. 4 is another schematic flowchart of a route determining method according to an embodiment of this application;

    FIG. 5 is a schematic structural block diagram of a route determining apparatus 300 according to an embodiment of this application;

    FIG. 6 is a schematic structural block diagram of a route determining apparatus 400 according to an embodiment of this application;

    FIG. 7 is a schematic structural block diagram of a route determining apparatus 500 according to an embodiment of this application;

    FIG. 8 is a schematic structural block diagram of a route determining device 600 according to an embodiment of this application; and

    FIG. 9 is a schematic diagram of a route determining system according to an embodiment of this application.


    DESCRIPTION OF EMBODIMENTS



    [0047] The following first describes the prior art for ease of understanding the technical solutions in embodiments of this application.

    [0048] FIG. 1 is a schematic diagram of a system that implements service flow routing and load balancing in an existing mobile broadband network technology. A policy and charging rules function (Policy and Charging Rules Function, PCRF for short) is responsible for a service chain decision function, that is, determines a type of service node that needs to process a type of service flow, and how to classify service flows. A policy and charging enforcement function (Policy and Charging Enforcement Function, PCEF for short) is configured to: recognize and classify service flows according to a classification rule formulated by the PCRF, and insert, into packets of the service flows, identifiers used to indicate service flow types. A software defined networking (Software Defined Network, SDN for short) controller (controller) is responsible for obtaining information about each service node and determining a service flow routing policy according to the information about the service node, that is, obtaining service nodes of each type and respective load statuses of the multiple service nodes of each type, and then determining a current service bearer proportion of each service node. An SDN switch is responsible for forwarding the packets of the service flows according to the routing policy determined by the SDN controller.

    [0049] In the prior art, load balancing is performed on a service node in the following two manners.

    [0050] Manner 1: The SDN controller obtains information such as a location (a corresponding switch and port), an address, a capacity, and dynamic load of each service node, and 5-tuple information and a service chain identifier (Identity, ID for short), that is, an identifier that indicates a service chain type, of each service flow.

    [0051] Then the SDN controller generates one flow entry for each service flow according to the information about each service node, determines a service node that each service flow needs to pass through, and sends the generated flow entry to the SDN switch. The SDN switch forwards a packet of the service flow according to the flow entry.

    [0052] In the manner 1, the SDN controller generates one flow entry for each service flow. Consequently, a quantity of flow entries is quite large, signaling overheads are extremely high, and efficiency is relatively low.

    [0053] Manner 2: Multiple service chain IDs (for example, 1 to 1000) are allocated to one type of service flows, and when classifying and identifying service flows, the PCEF may allocate different service chain IDs to a same type of service flows.

    [0054] The SDN controller obtains information such as a location (a corresponding switch and port), an address, a capacity, and dynamic load of each service node, and determines a service flow routing policy according to the information about the service node. For example, there are two service nodes of a type 1, and a capacity ratio between the two service nodes is 1:1. In this case, the SDN controller delivers a flow entry to the SDN switch, to instruct the SDN switch to route packets of service flows with service chain IDs 1 to 500 to a service node 1, and route packets of service flows with service chain IDs 501 to 1000 to a service node 2.

    [0055] In the manner 2, load balancing is not performed on a service node in a fully flexible manner. For example, if two more service nodes are newly added to the service nodes of the type 1, and capacities of the two service nodes are the same as those of the foregoing two service nodes, packets of service flows with service chain IDs 251 to 500 and 751 to 1000 may be separately transferred to the two new service nodes. However, in the manner 2, such a routing policy change can be performed only after a packet of an existing service flow is forwarded.

    [0056] Therefore, the prior art does not provide an efficient and flexible method for implementing service node load balancing. To resolve this problem, the embodiments of this application provide a route determining method, and a corresponding apparatus and system. The following describes technical solutions of the present invention in detail by using the accompanying drawings and specific embodiments. It should be understood that the embodiments of this application and specific features in the embodiments are intended to describe the technical solutions of the present invention in detail rather than limit the technical solutions of the present invention. In the case of no conflict, the embodiments of this application and the technical features in the embodiments may be mutually combined.

    [0057] FIG. 2A and FIG. 2B are a schematic flowchart of a route determining method according to this application. The method includes the following steps.

    [0058] Step 101: A controller obtains first current load of each service node existing at a first moment, and generates a first flow entry according to the obtained first current load of each service node, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node.

    [0059] Step 102: The controller sends the first flow entry to a switch.

    [0060] Step 103: The switch receives the first flow entry.

    [0061] Step 104: The controller obtains second current load of each service node existing at a second moment, and generates a second flow entry according to the second current load of each service node, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period.

    [0062] Step 105: The controller sends the second flow entry to the switch.

    [0063] Step 106: The switch receives the second flow entry.

    [0064] Step 107: The switch receives a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow.

    [0065] Step 108: The switch determines a time period that includes the first timestamp from the first effective time period and the second effective time period.

    [0066] Step 109: The switch determines, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet.

    [0067] Step 110: The switch sends the first packet to the determined service node.

    [0068] Specifically, the controller separately obtains information about service nodes existing at the first moment and the second moment. The information about the service nodes includes a location, an address, a capacity, and a current load proportion of each service node. The service nodes at the first moment may not be the same as the service nodes at the second moment. For example, two service nodes of a type 1 are newly added to a system at the second moment, and the service nodes at the second moment include the service nodes at the first moment and the two newly added service nodes.

    [0069] The controller separately generates the flow entries according to the information about the service nodes existing at the first moment and the second moment. The flow entries include the route mapping relationships of the packet of the first type of service flow and the load proportion of each service node. The first route mapping relationship in the first flow entry and the second route mapping relationship in the second flow entry may be the same, or may be different. This specifically depends on whether service nodes of each type are increased or decreased.

    [0070] For example, in the first route mapping relationship at the first moment, the packet that is of the first type of service flow and that is sent by a classifier needs to be sent to service nodes of the type 1, and the service nodes of the type 1 include a first service node and a second service node. At the second moment, one service node of the type 1, that is, a third service node is newly added. Therefore, in the second route mapping relationship, the packet that is of the first type of service flow and that is sent by the classifier is mapped to the first service node, the second service node, and the third service node. That is, a route mapping relationship may also change because of an increase or a decrease of service nodes.

    [0071] In this embodiment of this application, the first load proportion of the service node is different from the second load proportion of the service node. This includes the following two cases.

    [0072] First case: When a route mapping relationship changes because of an increase or a decrease of service nodes, a load proportion of each service node changes.

    [0073] Second case: When a route mapping relationship does not change, the controller adjusts a load proportion of each service node because current load of the service node is imbalanced. The current load of the service node includes a remaining capacity (a current bearable service volume) and a current load proportion of the service node. That the current load is imbalanced means that the remaining capacity of the service node does not match the current load proportion of the service node.

    [0074] For example, in the first route mapping relationship and the second route mapping relationship, all packets of the first type of service flow that are sent by the classifier are mapped to the first service node and the second service node, and remaining capacities of the first service node and the second service node are the same at the first moment. Therefore, the controller sets both first load proportions of the first service node and the second service node to 50%. At the second moment, a remaining capacity of the first service node is twice a remaining capacity of the second service node, and both current load proportions of the first service node and the second service node are 50%. The first service node with a higher service bearer capability has a same load proportion as the second service node with a lower bearer capability, and current load of the first service node and current load of the second service node are imbalanced. To implement load balancing, the controller may adjust a second load proportion of the first service node to 67%, and adjust a second load proportion of the second service node to 33%.

    [0075] Therefore, in this embodiment of this application, at the second moment, when the controller detects a service chain decision change or an increase or a decrease of service nodes, or when the current load of the service node is imbalanced, the controller generates the second flow entry according to the current load of each service node existing at the second moment. The second flow entry includes the second load proportion of each service node.

    [0076] In addition, because the controller has generated the first flow entry at the first moment, the second effective time period included in the second flow entry is different from the first effective time period of the first flow entry, to avoid a conflict. A start time of the second flow entry may be the second moment, or may be a time determined by increasing the second moment by one allowance.

    [0077] In addition, the switch separately receives the first flow entry and the second flow entry sent by the controller, and determines a route for the first type of service flow according to the first flow entry or the second flow entry.

    [0078] First, the switch receives the first packet of the service flow. The first packet may be from the classifier in the system, for example, a PCEF, a gateway (Gateway), or a traffic detection function (Traffic Detection Function, TDF for short). Alternatively, the first packet may be from a service node. For example, according to a service chain decision, the packet of the first type of service flow needs to successively pass through a service node that functions as a firewall and a second service node used for cache and acceleration. After passing through the service node that functions as a firewall, the packet needs to be returned to the switch and then forwarded, by using the switch, to the service node used for cache and acceleration.

    [0079] In this embodiment of this application, the first packet includes the service chain identifier and the first timestamp. The service chain identifier and the first timestamp may be added by the classifier in the system, and the first timestamp may indicate the creation time of the service flow, that is, a time at which the classifier receives a head packet of the first service flow.

    [0080] The switch determines, according to the service chain identifier included in the first packet, that the first packet is a packet of the first type of service flow, and needs to determine a route for the first packet according to the first flow entry or the second flow entry. The switch further determines whether the first timestamp included in the first packet is included in the first effective time period or the second effective time period. If the first effective time period includes a time corresponding to the first timestamp, the route for the first packet is determined according to the first flow entry. If the second effective time period includes a time corresponding to the first timestamp, the route for the first packet is determined according to the second flow entry.

    [0081] Actually, the switch may further receive a third flow entry for a routing policy of the service flow of the first service chain. A third effective time period of the third flow entry is different from both the first effective time period and the second effective time period. That is, the switch may receive and store multiple flow entries for a routing policy of service flows of a same service chain, and all effective time periods of the multiple flow entries are different from each other. After receiving a packet of the service flow of the first service chain, the switch selects an applicable flow entry according to a timestamp in the packet.

    [0082] In the foregoing technical solution in this embodiment of this application, when detecting that the load of the service node is imbalanced, the controller can immediately generate the second flow entry according to the current load of the service node, to instruct the switch to determine, according to the second route mapping relationship and the second load proportion, a route for a packet including a timestamp later than the start time of the second flow entry, so as to implement service node load balancing. Compared with the foregoing manner 1 in the prior art, in the route determining method provided in this embodiment of this application, a flow entry is generated for each type of service flows instead of each service flow, so that system overheads are greatly reduced. Compared with the foregoing manner 2 in the prior art, in the route determining method provided in this embodiment of this application, a route determining policy can be adjusted in a timely manner according to a change of the current load of the service node, instead of being adjusted after an existing packet is processed. This is more efficient and flexible.

    [0083] It should be noted that the controller may be an SDN controller in an SDN network architecture, and the switch may be an SDN switch in the SDN network architecture. If the SDN network architecture is not used, step 101, step 102, step 104, and step 105 may be performed by a network element that has a function the same as or similar to that of the SDN controller in the SDN network architecture, and step 103, and step 106 to step 110 may be performed by a network element that has a function the same as or similar to that of the SDN switch in the SDN network architecture. In addition, in another network architecture, alternatively, functions of the controller and the switch may be integrated into one network element.

    [0084] Optionally, in this embodiment of this application, after the switch receives the second flow entry in step 106, the method further includes the following steps:

    determining whether a start time of the second effective time period precedes an end time of the first effective time period; and

    if the start time of the second effective time period precedes the end time of the first effective time period, changing the end time of the first effective time period to the start time of the second effective time period. Specifically, a flow entry delivered by the controller to the switch may further include an effective time period field used to indicate an effective time of the flow entry. If an effective time period of a subsequently received (generated) flow entry overlaps an effective time period of a previously received (generated) flow entry, a route for a packet including a timestamp that falls within an overlapping time period is determined by using a policy defined in the subsequent flow entry, to respond to a latest routing policy and quickly implement load balancing.



    [0085] Actually, during generation of a flow entry, an effective time period field in the flow entry may be default, or may have a blank value, to indicate that the flow entry is always effective.

    [0086] Optionally, in this embodiment of this application, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0087] Referring to FIG. 3, step 109 in which the switch specifically determines, according to the first flow entry, the service node for receiving the first packet includes the following steps.

    [0088] Step 1091: Determine at least two service nodes according to an input node of the first packet and the first route mapping relationship.

    [0089] Step 1092: The switch generates a load balancing factor according to the first timestamp.

    [0090] Step 1093: The switch determines, from the at least two service nodes according to respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.

    [0091] Specifically, the route mapping relationship included in the flow entry is a mapping relationship between an input node and an output node of the packet of the first type of service flow, that is, indicates a service node to which the switch needs to forward a packet received from a first input node. The load proportions of the service nodes that are included in the flow entry are respective service bearer proportions of the at least two service nodes determined according to the route mapping relationship.

    [0092] For example, a possible form of the flow entry may be:

    "Match: InPort=0, SC-ID=1, T1≤TimeStamp<T2.

    Output: OutPort=1, Ratio=50%; OutPort=2, Ratio=50%."



    [0093] The Match paragraph includes information about a packet and a flow entry. InPort is a port identifier of an input node, SC-ID is a service chain identifier, and is also referred to as a service chain ID, and the TimeStamp field indicates that an effective time period of the flow entry is [T1, T2). As described above, the TimeStamp field may be default. The Output paragraph includes information about an output node. OutPort is an identifier of an output port, and Ratio is a load proportion corresponding to the output port. The foregoing flow entry indicates that packets that are input from a port 0 are sent to a port 1 or a port 2, and load proportions of the port 1 and the port 2 are the same, that is, the packets are equally distributed to the port 1 and the port 2.

    [0094] In step 1091, the switch obtains an identifier of the input node of the first packet, and determines, according to the identifier of the input node and the route mapping relationship in the first flow entry, the at least two service nodes that may be used for receiving the first packet.

    [0095] Then the switch generates the load balancing factor according to the first timestamp in the first packet, and determines, from the at least two service nodes (the service nodes determined in step 1091) with reference to the respective load proportions of the at least two service nodes in the first flow entry and the load balancing factor, the first service node for receiving the first packet.

    [0096] During specific implementation, step 1093 includes the following manners.

    [0097] Manner 1: The switch calculates a hash value of the timestamp, uses the calculated hash value as the load balancing factor, and determines, with reference to the hash value and the load proportions of the service nodes, the service node for receiving the packet.

    [0098] For example, two possible output service nodes determined according to the first route mapping relationship are a service node 1 and a service node 2, and load proportions of the two service nodes are 50%. In this case, after the hash value is calculated according to the timestamp, if the hash value is an odd number, the packet is sent to the service node 1. If the hash value is an even number, the packet is sent to the service node 2. Alternatively, an intermediate value is determined. If the hash value is greater than the intermediate value, the packet is sent to the service node 1. If the hash value is not greater than the intermediate value, the packet is sent to the service node 2.

    [0099] Manner 2: The switch may alternatively use the timestamp as the load balancing factor, and determine, according to the timestamp and the load proportions of the service nodes, the service node for receiving the packet.

    [0100] Still in an example in which the load proportions of both the service node 1 and the service node 2 are 50%, if the last digit of the timestamp is an odd number, the packet is sent to the service node 1. If the last digit of the timestamp is an even number, the packet is sent to the service node 2. Alternatively, a packet including a timestamp that within a range [T1, T1+T) is sent to the service node 1, a packet including a timestamp that falls within a range [T1+T, T1+2T) is sent to the service node 2, a packet including a timestamp that falls within a range [T1+2T, T1+3T) is sent to the service node 1, and so on.

    [0101] Manner 3: The switch generates the load balancing factor according to information unrelated to the timestamp, and determines, according to the load balancing factor and the load proportions of the service nodes, the service node for receiving the packet.

    [0102] For example, the switch consecutively numbers a first type of service flows in ascending order, and a number corresponding to each service flow is a load balancing factor of a packet of the service flow. For example, the switch numbers the first received first type of service flow 1, numbers the second received first type of service flow 2, ..., and numbers the tenth received first type of service flow 10, and then numbers service flows from 1 again. Still in an example in which the load proportions of both the service node 1 and the service node 2 are 50%, a packet numbered an odd number may be sent to the service node 1, and a packet numbered an even number may be sent to the service node 2. Alternatively, a packet whose number falls within a range [1, 5] may be sent to the service node 1, and a packet whose number falls within a range [6, 10] may be sent to the service node 2.

    [0103] It should be noted that, only several possible implementations of determining, according to the load balancing factor and the load proportions of the service nodes, the service node for receiving the packet are listed above, and persons skilled in the art may determine, in another possible implementation in the prior art, the service node for receiving the packet. No further examples are listed herein one by one in this embodiment of this application.

    [0104] Optionally, in this embodiment of this application, the first packet includes the load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0105] Step 109 in which the switch specifically determines, according to the first flow entry, the service node for receiving the first packet includes the following steps.

    [0106] Step 1091: The switch determines at least two service nodes according to an input node of the first packet and the first route mapping relationship.

    [0107] Step 1093: The switch determines, from the at least two service nodes according to respective first load proportions of the at least two service nodes and a load balancing factor, a first service node for receiving the first packet.

    [0108] Specifically, when identifying packets of service flows, the classifier may calculate a load balancing factor of a packet, and adds the load balancing factor to the packet. The switch directly obtains the load balancing factor from the packet without calculating the load balancing factor, and determines the service node for receiving the packet. For a manner in which the classifier calculates the load balancing factor and an implementation of step 1093, refer to the foregoing manner 1, manner 2, and manner 3, and details are not described herein again.

    [0109] In the foregoing technical solution, the load balancing factor is added to the packet, so that the switch does not need to calculate the load balancing factor. This reduces load on the switch.

    [0110] It should be noted that, in the foregoing two implementations of step 109, the first load proportion of the service node that is included in the first flow entry may be represented in a form of the "Ratio" field in the foregoing example, that is, a corresponding service bearer proportion of the service node. The first load proportion of the service node may be directly a range of a load balancing factor corresponding to the service node. For example, a first load proportion of the service node 1 may be represented as "LB≤Max/2", and a first load proportion of the service node 2 may be represented as "Max/2<LB≤SMax". LB is a load balancing factor.

    [0111] In addition, in the technical solution in this embodiment of this application, there may be multiple switches. After being processed by the first service node, the first packet needs to be returned to the switch and forwarded by the switch to a next service node. A switch that sends the first packet to the first service node and a switch that receives the packet sent by the first service node may be the same, or may be different. In this embodiment of this application, if an algorithm of the load balancing factor is uniform, packet route consistency can be ensured even if a switch is switched.

    [0112] Optionally, in this embodiment of this application, the first packet is the head packet of the first service flow. After step 1093 in which the switch determines, from the at least two service nodes, the first service node for receiving the first packet, the method further includes the following steps.

    [0113] Step 111: The switch receives a second packet of the first service flow.

    [0114] Step 112: The switch sends the second packet to the first service node.

    [0115] Specifically, because all packets of the first service flow have a same forwarding path, after determining that a service node for receiving the head packet of the first service flow is the first service node, the switch directly sends a packet subsequent to the head packet of the first service flow to the first service node, without determining routes for packets one by one according to a load proportion and a route mapping relationship.

    [0116] During specific implementation, after determining that the service node for receiving the head packet of the first service flow is the first service node, the switch may generate a new flow entry, and defines, in the new flow entry, that a packet of the first service flow is to be sent to the first service node. After receiving the subsequent packet of the first service flow, the switch may directly forward the packet to the first service node according to the new flow entry, without performing step 1093 to determine a next service node for receiving the packet.

    [0117] Step 111 and step 112 are applicable not only to the technical solution of determining a route for the packet when the first packet does not include the load balancing factor, but also to the technical solution of determining a route for the packet when the first packet includes the load balancing factor. The packet subsequent to the head packet of the first service flow is directly forwarded in a routing manner of the head packet of the first service flow, without a need to determine routes for packets of the service flow one by one by means of calculation, so that operation load on the switch is greatly reduced.

    [0118] Optionally, in this embodiment of this application, referring to FIG. 4, after step 1093 in which the switch determines, from the at least two service nodes, the first service node for receiving the first packet, the method further includes the following step.

    [0119] Step 113: The switch starts a timeout timer.

    [0120] Before step 112 in which the switch sends the second packet to the first service node, the method further includes the following step.

    [0121] Step 114: The switch determines that the timeout timer does not expire.

    [0122] Specifically, when the packet is directly forwarded in the manner in step 111 to step 113, the timeout timer is set. Before the timeout timer expires, a non-head packet of the first service flow is forwarded by performing step 111 and step 112.

    [0123] After the timeout timer expires, the packet may be forwarded in the following manner.

    [0124] First manner: At a moment at which the timeout timer expires, it is determined whether there is a to-be-processed (to-be-forwarded) packet of the first service flow. If there is a to-be-processed (to-be-forwarded) packet, the timer is reset, and the timeout timer is canceled until there is no packet of the first service flow when the timeout timer expires.

    [0125] Second manner: For a first packet after the timeout timer expires, a service node for receiving the first packet is determined in a manner in step 1093, then a new timeout timer is created, and step 114 and step 112 are performed. The previous timeout timer may be canceled.

    [0126] The foregoing technical solution is intended to achieve the following purpose: As described above, step 111 and step 112 are usually performed by using a newly created flow entry, and in the foregoing solution in which the timeout timer is used, the newly created flow entry may be aged, and after packets of the first service flow are processed, the newly created flow entry is deleted, so that a quantity of flow entries is reduced, and overheads are reduced.

    [0127] Based on a same inventive concept, an embodiment of this application further provides a route determining method. Still referring to FIG. 2A and FIG. 2B, the method includes the following steps.

    [0128] Step 201: A classifier receives a first packet of a first service flow.

    [0129] Step 202: The classifier determines, according to a feature field included in the first packet, that the first service flow is a service flow of a first service chain.

    [0130] Step 203: The classifier adds a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow.

    [0131] Step 204: The classifier sends the first packet to a switch.

    [0132] Specifically, step 201 to step 204 may be performed by the classifier in a service chain architecture. The classifier may be specifically a PCEF, a gateway, a TDF, or the like. Actually, step 201 to step 204 may be performed by another network element that has a function the same as or similar to that of the classifier in the service chain architecture.

    [0133] In this embodiment of this application, the classifier classifies and identifies service flows according to a service chain decision policy. The service chain decision policy includes a rule for identifying the service flows, and the rule is generated by a PCRF in the service chain architecture, and is obtained by the classifier from the PCRF. Specifically, there may be a PUSH mode or a PULL mode. The PUSH mode means that the PCRF actively pushes a rule for classifying user-related service flows into service chains to the classifier, and the PULL mode means that after receiving a service flow initiated by a user, the classifier pulls a corresponding service chain decision policy from the PCRF.

    [0134] In the foregoing technical solution, when classifying and identifying packets of service flows, the classifier adds a timestamp to the packet of the service flow, to add a time attribute to the packet, so that the switch can use different routing policies for packets in different time periods. Therefore, service node load balancing is implemented flexibly and efficiently.

    [0135] In addition, in the foregoing technical solution in this embodiment of this application, a controller and the switch perform service flow service chain decision, service flow recognition, service flow identification, and service flow route decoupling, and implement a load balancing policy, without exchanging signaling with the PCRF or the classifier, so that network efficiency is improved.

    [0136] Optionally, in this embodiment of this application, before step 204, the method further includes the following step.

    [0137] Step 205: The classifier determines a load balancing factor of the first packet, and adds the load balancing factor to the first packet.

    [0138] Specifically, the classifier may calculate the load balancing factor of the packet according to the timestamp, or may generate the load balancing factor of the packet according to other information unrelated to the timestamp. An implementation thereof is similar to the manner 1, the manner 2, and the manner 3 in step 1093, and details are not described herein again.

    [0139] The classifier adds the load balancing factor to the packet, so that operation load on the switch can be reduced.

    [0140] Based on a same technical concept, an embodiment of this application further provides a route determining apparatus 300. Referring to FIG. 5, the apparatus 300 includes:

    a first receiving module 301, configured to: receive a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry; and receive a second flow entry, where the second flow entry includes a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period;

    a second receiving module 302, configured to receive a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    a first determining module 303, configured to determine a time period that includes the first timestamp from the first effective time period and the second effective time period;

    a second determining module 304, configured to determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and

    a sending module 305, configured to send the first packet to the determined service node.



    [0141] Optionally, in this embodiment of this application, the apparatus further includes:
    a change module 306, configured to: after the second flow entry is received, determine whether a start time of the second effective time period precedes an end time of the first effective time period; and when the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.

    [0142] Optionally, in this embodiment of this application, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0143] The second determining module 304 is specifically configured to:

    determine at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generate a load balancing factor according to the first timestamp; and

    determine, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0144] Optionally, in this embodiment of this application, the first packet further includes a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0145] The second determining module 304 is specifically configured to:

    determine at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determine, from the at least two service nodes according to the respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0146] Optionally, in this embodiment of this application, the first packet is a head packet of the first service flow.

    [0147] The second receiving module 302 is further configured to receive a second packet of the first service flow.

    [0148] The sending module 305 is further configured to send the second packet to the first service node.

    [0149] Optionally, in this embodiment of this application, the apparatus further includes:
    a third determining module 307, configured to: start a timeout timer after the first service node for receiving the first packet is determined from the at least two service nodes; and before the second packet is sent to the first service node, determine that the timeout timer does not expire.

    [0150] The apparatus 300 in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the apparatus 300 in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0151] Based on a same inventive concept, an embodiment of this application further provides a route determining apparatus 400. Referring to FIG. 6, the apparatus 400 includes:

    a receiving module 401, configured to receive a first packet of a first service flow;

    a determining module 402, configured to determine, according to a feature field included in the packet, that the first service flow is a service flow of a first service chain;

    an identification module 403, configured to add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and

    a sending module 404, configured to send the first packet to a switch.



    [0152] Optionally, in this embodiment of this application, the identification module 403 is further configured to: determine a load balancing factor of the first packet, and add the load balancing factor to the first packet.

    [0153] The apparatus 400 in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the apparatus 400 in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0154] Based on a same inventive concept, an embodiment of this application further provides a route determining apparatus 500. Referring to FIG. 7, the apparatus 500 includes:

    a first obtaining module 501, configured to obtain first current load of each service node existing at a first moment;

    a first generation module 502, configured to generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node;

    a sending module 503, configured to send the first flow entry to a switch;

    a second obtaining module 504, configured to obtain second current load of each service node existing at a second moment; and

    a second generation module 505, configured to generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period.



    [0155] The sending module 503 is further configured to send the second flow entry to the switch.

    [0156] The apparatus 500 in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the apparatus 500 in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0157] Based on a same inventive concept, an embodiment of this application further provides a route determining device 600. Referring to FIG. 8, the device 600 includes:

    a receiver 601, configured to: receive a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry; receive a second flow entry, where the second flow entry includes a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period; and receive a first packet of a first service flow, where the first packet includes a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    a processor 602, configured to: determine a time period that includes the first timestamp from the first effective time period and the second effective time period; and determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and

    a transmitter 603, configured to send the first packet to the determined service node.



    [0158] Optionally, in this embodiment of this application, the processor 602 is further configured to: start a timeout timer after a first service node for receiving the first packet is determined from at least two service nodes; and before the transmitter sends a second packet to the first service node, determine that the timeout timer does not expire.

    [0159] Optionally, in this embodiment of this application, the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0160] That the processor 602 is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship;

    generating a load balancing factor according to the first timestamp; and

    determining, from the at least two service nodes according to respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0161] Optionally, in this embodiment of this application, the first packet further includes a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that includes the first timestamp is the first time period.

    [0162] That the processor 602 is configured to determine, according to the first flow entry, the service node for receiving the first packet includes:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship; and

    determining, from the at least two service nodes according to respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet.



    [0163] Optionally, in this embodiment of this application, the first packet is a head packet of the first service flow.

    [0164] The receiver 601 is configured to receive a second packet of the first service flow.

    [0165] The transmitter is further configured to send the second packet to the first service node.

    [0166] Optionally, in this embodiment of this application, the processor 602 is further configured to: start a timeout timer after the first service node for receiving the first packet is determined from the at least two service nodes; and before the transmitter sends the second packet to the first service node, determine that the timeout timer does not expire.

    [0167] Optionally, in this embodiment of this application, the device 600 further includes a memory 604, configured to store the received packet and flow entries.

    [0168] The device 600 in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the device 600 in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0169] Based on a same inventive concept, an embodiment of this application further provides a route determining device, and the device includes:

    a receiver, configured to receive a first packet of a first service flow;

    a processor, configured to: determine, according to a feature field included in the first packet, that the first service flow is a service flow of a first service chain; and add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and

    a transmitter, configured to send the first packet to a switch.



    [0170] Optionally, in this embodiment of this application, the processor is further configured to: determine a load balancing factor of the first packet, and add the load balancing factor to the first packet.

    [0171] Optionally, in this embodiment of this application, the device further includes a memory, configured to store the received packet.

    [0172] The device in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the device in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0173] Based on a same inventive concept, an embodiment of this application further provides a route determining device, and the device includes:

    a receiver, configured to obtain first current load of each service node existing at a first moment;

    a processor, configured to generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node; and

    a transmitter, configured to send the first flow entry to a switch.



    [0174] The receiver is further configured to obtain second current load of each service node existing at a second moment.

    [0175] The processor is further configured to generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period.

    [0176] The transmitter is further configured to send the second flow entry to the switch.

    [0177] Optionally, in this embodiment of this application, the device further includes a memory, configured to store received load information of the service node and the generated flow entries.

    [0178] The device in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the device in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0179] It should be noted that the processor may be one processing chip, or may be a collective term of multiple processing elements. For example, the processor may be a central processing unit (English: Central Processing Unit, CPU for short), or may be an application-specific integrated circuit (English: Application Specific Intergrated Circuit, ASIC for short), or may be one or more integrated circuits configured to implement this embodiment of the present invention, for example, one or more microprocessors (English: digital singnal processor, DSP for short) or one or more field programmable gate arrays (English: Field Programmable Gate Array, FPGA for short).

    [0180] In addition, the receiver and the transmitter can be integrated together.

    [0181] Based on a same inventive concept, an embodiment of this application further provides a route determining system. Referring to FIG. 9, the system includes:

    a controller 701, configured to: obtain first current load of each service node existing at a first moment; generate a first flow entry, where the first flow entry includes a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node; send the first flow entry to a switch; obtain second current load of each service node existing at a second moment; generate a second flow entry, where the second flow entry includes a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second load proportion is generated according to the second current load of each service node; and send the second flow entry to the switch, where the second effective time period is different from the first effective time period;

    a classifier 702, configured to: receive a first packet of a first service flow; determine, according to a feature field included in the first packet, that the first service flow is a service flow of the first service chain; add a first timestamp of the first packet and a first service chain identifier to the first packet, where the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and send the first packet to the switch; and

    the switch 703, configured to: receive the first flow entry; receive the second flow entry; receive the first packet of the first service flow; determine a time period that includes the first timestamp from the first effective time period and the second effective time period; determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and send the first packet to the determined service node.



    [0182] Optionally, in this embodiment of this application, the switch 703 is further configured to: after receiving the second flow entry,
    determine whether a start time of the second effective time period precedes an end time of the first effective time period; and
    if the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.

    [0183] Optionally, in this embodiment of this application, the first packet is a head packet of the first service flow.

    [0184] The switch 703 is further configured to: receive a second packet of the first service flow; and send the second packet to the first service node.

    [0185] The system in this embodiment and the methods corresponding to FIG. 2A and FIG. 2B to FIG. 4 are two aspects based on a same inventive concept, and an implementation process of the methods is described in detail above. Therefore, persons skilled in the art may clearly understand a structure and an implementation process of the system in this embodiment according to the foregoing description. For conciseness of this specification, details are not described herein again.

    [0186] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:
    In the embodiments of this application, the switch receives the first flow entry and the second flow entry, and determines, according to the timestamp included in the first packet, whether to use the first flow entry or the second flow entry to determine a route for the first packet. Therefore, when a new flow entry is generated because a load status of a service node changes, the switch can immediately use the new flow entry to forward a packet of a newly created service flow, and continue to forward a packet of a previously created service flow based on a previous flow entry. Different from the prior art in which a packet is forwarded according to a new flow entry only after an existing packet is processed, in the solution provided in the embodiments of this application, the switch can adjust a routing policy in a timely manner and implement load balancing more flexibly and quickly. In addition, in the solution provided in the embodiments of this application, a flow entry is generated for a same type of service flows instead of each service flow. Therefore, there are few flow entries, system overheads are relatively low, and efficiency is relatively high.

    [0187] Persons skilled in the art should understand that the embodiments of the present invention may be provided as a method, a system, or a computer program product. Therefore, the present invention may use a form of hardware only embodiments, software only embodiments, or embodiments with a combination of software and hardware. Moreover, the present invention may use a form of a computer program product that is implemented on one or more computer-usable storage media (including but not limited to a disk memory, a CD-ROM, an optical memory, and the like) that include computer-usable program code.

    [0188] The present invention is described with reference to the flowcharts and/or block diagrams of the method, the device (system), and the computer program product according to the embodiments of this application. It should be understood that computer program instructions may be used to implement each process and/or each block in the flowcharts and/or the block diagrams and a combination of a process and/or a block in the flowcharts and/or the block diagrams. These computer program instructions may be provided for a general-purpose computer, a dedicated computer, an embedded processor, or a processor of any other programmable data processing device to generate a machine, so that the instructions executed by the computer or the processor of any other programmable data processing device generate an apparatus for implementing a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

    [0189] These computer program instructions may be stored in a computer readable memory that can instruct the computer or any other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory generate an artifact that includes an instruction apparatus. The instruction apparatus implements a specific function in one or more processes in the flowcharts and/or in one or more blocks in the block diagrams.

    [0190] Although some preferred embodiments of the present invention have been described, persons skilled in the art can make changes and modifications to these embodiments provided that they learn the basic inventive concept. Therefore, the following claims are intended to be construed as to cover the preferred embodiments and all changes and modifications falling within the scope of the present invention.

    [0191] Obviously, persons skilled in the art can make various modifications and variations to the present invention without departing from the scope of the present invention. The present invention is intended to cover these modifications and variations provided that they fall within the scope of protection defined by the following claims and their equivalent technologies.


    Claims

    1. A route determining method, comprising:

    receiving a first flow entry (103), wherein the first flow entry comprises a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry;

    receiving a second flow entry (106), wherein the second flow entry comprises a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second effective time period is different from the first effective time period;

    receiving a first packet of a first service flow (107), wherein the first packet comprises a first timestamp of the first service flow and a first service chain identifier, the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow;

    determining, from the first effective time period and the second effective time period, a time period that includes the first timestamp (108);

    determining, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet (109); and

    sending the first packet to the determined service node (110).


     
    2. The method according to claim 1, after the receiving a second flow entry (106), further comprising:

    determining whether a start time of the second effective time period precedes an end time of the first effective time period; and

    if the start time of the second effective time period precedes the end time of the first effective time period, changing the end time of the first effective time period to the start time of the second effective time period.


     
    3. The method according to claim 1 or 2, wherein the time period that is determined from the first effective time period and the second effective time period and that comprises the first timestamp is the first effective time period; and
    the determining, according to the first flow entry, a service node for receiving the first packet (109) comprises:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship (1091);

    generating a load balancing factor according to the first timestamp (1092); and

    determining, from the at least two service nodes according to respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet (1093).


     
    4. The method according to claim 1 or 2, wherein the first packet further comprises a load balancing factor, and the time period that is determined from the first effective time period and the second effective time period and that comprises the first timestamp is the first effective time period; and
    the determining, according to the first flow entry, a service node for receiving the first packet (109) comprises:

    determining at least two service nodes according to an input node of the first packet and the first route mapping relationship (1091); and

    determining, from the at least two service nodes according to respective first load proportions of the at least two service nodes and the load balancing factor, a first service node for receiving the first packet (1093).


     
    5. The method according to claim 3 or 4, wherein the first packet is a head packet of the first service flow, and after the determining, from the at least two service nodes, a first service node for receiving the first packet (1093), the method further comprises:

    receiving a second packet of the first service flow (111); and

    sending the second packet to the first service node (112).


     
    6. The method according to claim 5, after the determining, from the at least two service nodes, a first service node for receiving the first packet (1093), further comprising:

    starting a timeout timer (113); and

    before the sending the second packet to the first service node (112), further comprising:
    determining that the timeout timer does not expire (114).


     
    7. A route determining method, comprising:

    receiving a first packet of a first service flow (201);

    determining, according to a feature field comprised in the first packet, that the first service flow is a service flow of a first service chain (202);

    adding a first timestamp of the first packet and a first service chain identifier to the first packet, wherein the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow (203); and

    sending the first packet to a switch (204) adapted to perform the method steps of claim 1.


     
    8. The method according to claim 7, before the sending the first packet to a switch (203), further comprising:
    determining a load balancing factor of the first packet, and adding the load balancing factor to the first packet.
     
    9. A route determining method, comprising:

    obtaining first current load of each service node existing at a first moment (101);

    generating a first flow entry, wherein the first flow entry comprises a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node (101);

    sending the first flow entry to a switch (102);

    obtaining second current load of each service node existing at a second moment (104);

    generating a second flow entry, wherein the second flow entry comprises a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period (104); and

    sending the second flow entry to the switch adapted to perform the method step of claim 1.


     
    10. A route determining apparatus (300), configured to perform the method according to any one of claims 1-6.
     
    11. A route determining apparatus (400), configured to perform the method according to claim 7 or 8.
     
    12. A route determining apparatus (500), comprising:

    a first obtaining module (501), configured to obtain first current load of each service node existing at a first moment;

    a first generation module (502), configured to generate a first flow entry, wherein the first flow entry comprises a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node;

    a sending module (503), configured to send the first flow entry to a switch;

    a second obtaining module (504), configured to obtain second current load of each service node existing at a second moment; and

    a second generation module (505), configured to generate a second flow entry, wherein the second flow entry comprises a second route mapping relationship of the service flow of the first service chain, a second load proportion of each service node, and a second effective time period of the second flow entry, the second load proportion is generated according to the second current load of each service node, and the second effective time period is different from the first effective time period, wherein

    the sending module is further configured to send the second flow entry to the switch adapted to perform the method according to claim 1.


     
    13. A route determining system, comprising:

    a controller (701), configured to: obtain first current load of each service node existing at a first moment; generate a first flow entry, wherein the first flow entry comprises a first route mapping relationship of a service flow of a first service chain, a first load proportion of each service node, and a first effective time period of the first flow entry, and the first load proportion is generated according to the first current load of each service node; send the first flow entry to a switch; obtain second current load of each service node existing at a second moment; generate a second flow entry, wherein the second flow entry comprises a second route mapping relationship of a packet of the first type of service flow, a second load proportion of each service node, and a second effective time period of the second flow entry, and the second load proportion is generated according to the second current load of each service node; and send the second flow entry to the switch, wherein the second effective time period is different from the first effective time period;

    a classifier (702), configured to: receive a first packet of a first service flow; determine, according to a feature field comprised in the first packet, that the first service flow is a service flow of the first service chain; add a first timestamp of the first packet and a first service chain identifier to the first packet, wherein the first service chain identifier indicates that the first service flow is a service flow of the first service chain, and the first timestamp indicates a creation time of the first service flow; and send the first packet to the switch; and

    the switch (703), configured to: receive the first flow entry; receive the second flow entry; receive the first packet of the first service flow; determine, from the first effective time period and the second effective time period, a time period that includes the first timestamp; determine, according to a flow entry corresponding to the determined time period, a service node for receiving the first packet; and send the first packet to the determined service node.


     
    14. The system according to claim 13, wherein the switch is further configured to: after receiving the second flow entry,
    determine whether a start time of the second effective time period precedes an end time of the first effective time period; and
    if the start time of the second effective time period precedes the end time of the first effective time period, change the end time of the first effective time period to the start time of the second effective time period.
     
    15. The system according to claim 13 or 14, wherein the first packet is a head packet of the first service flow; and
    the switch is further configured to: receive a second packet of the first service flow; and send the second packet to a first service node.
     


    Ansprüche

    1. Routenermittlungsverfahren, umfassend:

    Empfangen eines ersten Datenstromeintritts (103), wobei der erste Datenstromeintritt umfasst: eine erste Routenzuordnungsbeziehung eines Dienstdatenstroms einer ersten Dienstkette, einen ersten Lastanteil von jedem Dienstknoten und eine erste effektive Zeitdauer des ersten Datenstromeintritts;

    Empfangen eines zweiten Datenstromeintritts (106), wobei der zweite Datenstromeintritt umfasst: eine zweite Routenzuordnungsbeziehung des Dienstdatenstroms der ersten Dienstkette, einen zweiten Lastanteil von jedem Dienstknoten und eine zweite effektive Zeitdauer des zweiten Datenstromeintritts; und

    wobei die zweite effektive Zeitdauer verschieden von der ersten effektiven Zeitdauer ist;

    Empfangen eines ersten Datenpakets eines ersten Dienstdatenstroms (107), wobei das erste Datenpaket eine erste Zeitmarke des ersten Dienstdatenstroms und eine erste Dienstkettenkennung umfasst, wobei die erste Dienstkettenkennung anzeigt, dass der erste Dienstdatenstrom ein Dienstdatenstrom der ersten Dienstkette ist, und wobei die erste Zeitmarke einen Erstellungszeitpunkt des ersten Dienstdatenstroms anzeigt;

    Ermitteln aus der ersten effektiven Zeitdauer und der zweiten effektiven Zeitdauer einer Zeitdauer, welche die erste Zeitmarke enthält (108);

    Ermitteln gemäß einem Datenstromeintritt, welcher der ermittelten Zeitdauer entspricht, eines Dienstknotens zum Empfangen des ersten Datenpakets (109); und

    Senden des ersten Datenpakets an den ermittelten Dienstknoten (110).


     
    2. Verfahren nach Anspruch 1, das nach dem Empfangen eines zweiten Datenstromeintritts (106) außerdem umfasst:

    Ermitteln, ob ein Startzeitpunkt der zweiten effektiven Zeitdauer vor einem Endzeitpunkt der ersten effektiven Zeitdauer liegt; und

    wenn der Startzeitpunkt der zweiten effektiven Zeitdauer vor dem Endzeitpunkt der ersten effektiven Zeitdauer liegt, Ändern des Endzeitpunkts der ersten effektiven Zeitdauer auf den Startzeitpunkt der zweiten effektiven Zeitdauer.


     
    3. Verfahren nach Anspruch 1 oder 2, wobei die Zeitdauer, die aus der ersten effektiven Zeitdauer und der zweiten effektiven Zeitdauer ermittelt wird und welche die erste Zeitmarke umfasst, die erste effektive Zeitdauer ist; und
    wobei das Ermitteln gemäß dem ersten Datenstromeintritt eines Dienstknotens zum Empfangen des ersten Datenpakets (109) umfasst:

    Ermitteln von mindestens zwei Dienstknoten gemäß einem Eingangsknoten des ersten Datenpakets und der ersten Routenzuordnungsbeziehung (1091);

    Erzeugen eines Lastausgleichfaktors gemäß der ersten Zeitmarke (1092); und

    Ermitteln aus den mindestens zwei Dienstknoten gemäß den jeweiligen Lastanteilen der mindestens zwei Dienstknoten und dem Lastausgleichfaktor eines ersten Dienstknotens zum Empfangen des ersten Datenpakets (1093).


     
    4. Verfahren nach Anspruch 1 oder 2, wobei das erste Datenpaket außerdem einen Lastausgleichfaktor umfasst, und wobei die Zeitdauer, die aus der ersten effektiven Zeitdauer und der zweiten effektiven Zeitdauer ermittelt wird und welche die erste Zeitmarke umfasst, die erste effektive Zeitdauer ist; und
    wobei das Ermitteln gemäß dem ersten Datenstromeintritt eines Dienstknotens zum Empfangen des ersten Datenpakets (109) umfasst:

    Ermitteln von mindestens zwei Dienstknoten gemäß einem Eingangsknoten des ersten Datenpakets und der ersten Routenzuordnungsbeziehung (1091); und

    Ermitteln gemäß den jeweiligen Lastanteilen der mindestens zwei Dienstknoten und dem Lastausgleichfaktor, eines ersten Dienstknotens zum Empfangen des ersten Datenpakets aus den mindestens zwei Dienstknoten (1093).


     
    5. Verfahren nach Anspruch 3 oder 4, wobei das erste Datenpaket ein Kopfdatenpaket des ersten Dienstdatenstroms ist, und wobei das Verfahren nach dem Ermitteln eines ersten Dienstknotens zum Empfangen des ersten Datenpakets (1093) aus den mindestens zwei Dienstknoten außerdem umfasst:

    Empfangen eines zweiten Datenpakets des ersten Dienstdatenstroms (111), und

    Senden des zweiten Datenpakets an den ersten Dienstknoten (112).


     
    6. Verfahren nach Anspruch 5, das nach dem Ermitteln eines ersten Dienstknotens zum Empfangen des ersten Datenpakets (1093) aus den mindestens zwei Dienstknoten außerdem umfasst:

    Starten einer Zeitüberschreitungsmesseinheit (113); und

    wobei das Verfahren vor dem Senden des zweiten Datenpakets an den ersten Dienstknoten (112) außerdem umfasst:
    Ermitteln, dass die Zeitüberschreitungsmesseinheit nicht abgelaufen ist (114).


     
    7. Routenermittlungsverfahren, umfassend:

    Empfangen eines ersten Datenpakets eines ersten Dienstdatenstroms (201);

    Ermitteln gemäß einem Merkmalsfeld, das in dem ersten Datenpaket enthalten ist, dass der erste Dienstdatenstrom ein Dienstdatenstrom einer ersten Dienstkette ist (202);

    Hinzufügen einer ersten Zeitmarke des ersten Datenpakets und einer ersten Dienstkettenkennung zu dem ersten Datenpaket, wobei die erste Dienstkettenkennung anzeigt, dass der erste Dienstdatenstrom ein Dienstdatenstrom der ersten Dienstkette ist, und wobei die erste Zeitmarke einen Erstellungszeitpunkt des ersten Dienstdatenstroms anzeigt (203); und

    Senden des ersten Datenpakets an einen Verteiler (204), der geeignet ist zum Ausführen der Verfahrensschritte nach Anspruch 1.


     
    8. Verfahren nach Anspruch 7, das vor dem Senden des ersten Datenpakets an einen Verteiler (203) außerdem umfasst:
    Ermitteln eines Lastausgleichfaktors des ersten Datenpakets und Hinzufügen des Lastausgleichfaktors zu dem ersten Datenpaket.
     
    9. Routenermittlungsverfahren, umfassend:

    Erhalten einer an einem ersten Zeitpunkt (101) vorhandenen ersten aktuellen Last von jedem Dienstknoten;

    Erzeugen eines ersten Datenstromeintritts, wobei der erste Datenstromeintritt umfasst:

    eine erste Routenzuordnungsbeziehung eines Dienstdatenstroms einer ersten Dienstkette, einen ersten Lastanteil von jedem Dienstknoten und eine erste effektive Zeitdauer des ersten Datenstromeintritts, und wobei der erste Lastanteil gemäß der ersten aktuellen Last von jedem Dienstknoten erzeugt wird (101);

    Senden des ersten Datenstromeintritts an einen Verteiler (102);

    Erhalten einer an einem zweiten Zeitpunkt (104) vorhandenen zweiten aktuellen Last von jedem Dienstknoten;

    Erzeugen eines zweiten Datenstromeintritts, wobei der zweite Datenstromeintritt umfasst: eine zweite Routenzuordnungsbeziehung des Dienstdatenstroms der ersten Dienstkette, einen zweiten Lastanteil von jedem Dienstknoten und eine zweite effektive Zeitdauer des zweiten Datenstromeintritts, wobei der zweite Lastanteil gemäß der zweiten aktuellen Last von jedem Dienstknoten erzeugt wird, und wobei die zweite effektive Zeitdauer verschieden von der ersten effektiven Zeitdauer ist (104); und

    Senden des zweiten Datenstromeintritts an den Verteiler, der geeignet ist zum Ausführen der Verfahrensschritte nach Anspruch 1.


     
    10. Routenermittlungsvorrichtung (300), die konfiguriert ist zum Ausführen des Verfahrens nach einem der Ansprüche 1 bis 6.
     
    11. Routenermittlungsvorrichtung (400), die konfiguriert ist zum Ausführen des Verfahrens nach einem der Ansprüche 7 oder 8.
     
    12. Routenermittlungsvorrichtung (500), umfassend:

    ein erstes Erhaltemodul (501), das konfiguriert ist zum Erhalten einer ersten an einem ersten Zeitpunkt vorhandenen aktuellen Last von jedem Dienstknoten;

    ein erstes Erzeugungsmodul (502), das konfiguriert ist zum Erzeugen eines ersten Datenstromeintritts, wobei der erste Datenstromeintritt umfasst: eine erste Routenzuordnungsbeziehung eines Dienstdatenstroms einer ersten Dienstkette, einen ersten Lastanteil von jedem Dienstknoten und eine erste effektive Zeitdauer des ersten Datenstromeintritts, und wobei der erste Lastanteil gemäß der ersten aktuellen Last von jedem Dienstknoten erzeugt wird;

    ein Sendemodul (503), das konfiguriert ist zum Senden des ersten Datenstromeintritts an einen Verteiler;

    ein zweites Erhaltemodul (504), das konfiguriert ist zum Erhalten einer an einem zweiten Zeitpunkt vorhandenen zweiten aktuellen Last von jedem Dienstknoten; und

    ein zweites Erzeugungsmodul (505), das konfiguriert ist zum Erzeugen eines zweiten Datenstromeintritts, wobei der zweite Datenstromeintritt umfasst: eine zweite Routenzuordnungsbeziehung des Dienstdatenstroms der ersten Dienstkette, einen zweiten Lastanteil von jedem Dienstknoten und eine zweite effektive Zeitdauer des zweiten Datenstromeintritts, wobei der zweite Lastanteil gemäß der zweiten aktuellen Last von jedem Dienstknoten erzeugt wird, und wobei die zweite effektive Zeitdauer verschieden von der ersten effektiven Zeitdauer ist, wobei das Sendemodul außerdem konfiguriert ist zum Senden des zweiten Datenstromeintritts an den Verteiler, der geeignet ist zum Ausführen des Verfahrens nach Anspruch 1.


     
    13. Routenermittlungssystem, umfassend:

    eine Steuereinheit (701), die konfiguriert ist zum: Erhalten einer an einem ersten Zeitpunkt vorhandenen ersten aktuellen Last von jedem Dienstknoten; Erzeugen eines ersten Datenstromeintritts, wobei der erste Datenstromeintritt umfasst: eine erste Routenzuordnungsbeziehung eines Dienstdatenstroms einer ersten Dienstkette, einen ersten Lastanteil von jedem Dienstknoten und eine erste effektive Zeitdauer des ersten Datenstromeintritts, und wobei der erste Lastanteil gemäß der ersten aktuellen Last von jedem Dienstknoten erzeugt wird; Senden des ersten Datenstromeintritts an einen Verteiler; Erhalten einer an einem zweiten Zeitpunkt vorhandenen zweiten aktuellen Last von jedem Dienstknoten, die an einem zweiten Zeitpunkt vorhanden ist; Erzeugen eines zweiten Datenstromeintritts, wobei der zweite Datenstromeintritt umfasst: eine zweite Routenzuordnungsbeziehung eines Datenpakets des ersten Typs des Dienstdatenstroms, einen zweiten Lastanteil von jedem Dienstknoten und eine zweite effektive Zeitdauer des zweiten Datenstromeintritts, und wobei der zweite Lastanteil gemäß der zweiten aktuellen Last von jedem Dienstknoten erzeugt wird; und Senden des zweiten Datenstromeintritts an den Verteiler; wobei die zweite effektive Zeitdauer verschieden von der ersten effektiven Zeitdauer ist,

    eine Klassifizierungseinheit (702), die konfiguriert ist zum: Empfangen eines ersten Datenpakets eines ersten Dienstdatenstroms; Ermitteln gemäß einem Merkmalsfeld, das in dem ersten Datenpaket enthalten ist, dass der erste Dienstdatenstrom ein Dienstdatenstrom der ersten Dienstkette ist; Hinzufügen einer ersten Zeitmarke des ersten Datenpakets und einer ersten Dienstkettenkennung zu dem ersten Datenpaket, wobei die erste Dienstkettenkennung anzeigt, dass der erste Dienstdatenstrom ein Dienstdatenstrom der ersten Dienstkette ist, und wobei die erste Zeitmarke einen Erstellungszeitpunkt des ersten Dienstdatenstroms anzeigt; und Senden des ersten Datenpakets an den Verteiler; und

    wobei der Verteiler (703) konfiguriert ist zum: Empfangen des ersten Datenstromeintritts; Empfangen des zweiten Datenstromeintritts; Empfangen des ersten Datenpakets des ersten Dienstdatenstroms; Ermitteln aus der ersten effektiven Zeitdauer und der zweiten effektiven Zeitdauer einer Zeitdauer, welche die erste Zeitmarke enthält; Ermitteln gemäß einem Datenstromeintritt, welcher der ermittelten Zeitdauer entspricht, eines Dienstknotens zum Empfangen des ersten Datenpakets; und

    Senden des ersten Datenpakets an den ermittelten Dienstknoten.


     
    14. System nach Anspruch 13, wobei der Verteiler außerdem konfiguriert ist zum: nach dem Empfangen des zweiten Datenstromeintritts, Ermitteln, ob ein Startzeitpunkt der zweiten effektiven Zeitdauer vor einem Endzeitpunkt der ersten effektiven Zeitdauer liegt; und
    wenn der Startzeitpunkt der zweiten effektiven Zeitdauer vor dem Endzeitpunkt der ersten effektiven Zeitdauer liegt, Ändern des Endzeitpunkts der ersten effektiven Zeitdauer auf den Startzeitpunkt der zweiten effektiven Zeitdauer.
     
    15. System nach Anspruch 13 oder 14, wobei das erste Datenpaket ein Kopfdatenpaket des ersten Dienstdatenstroms ist; und
    wobei der Verteiler außerdem konfiguriert ist zum: Empfangen eines zweiten Datenpakets des ersten Dienstdatenstroms; und Senden des zweiten Datenpakets an einen ersten Dienstknoten.
     


    Revendications

    1. Procédé de détermination de voie d'acheminement, comprenant :

    la réception d'une première entrée de flux (103), la première entrée de flux comprenant une première relation de mappage de voie d'acheminement d'un flux de services d'une première chaîne de service, une première proportion de charge de chaque nœud de service et une première période de temps effective de la première entrée de flux ;

    la réception d'une deuxième entrée de flux (106), la deuxième entrée de flux comprenant une deuxième relation de mappage de voie d'acheminement du flux de services de la première chaîne de services, une deuxième proportion de charge de chaque nœud de service et une deuxième période de temps effective de la deuxième entrée de flux, et la deuxième période de temps effective étant différente de la première période de temps effective ;

    la réception d'un premier paquet d'un premier flux de services (107), le premier paquet comprenant un premier horodatage du premier flux de services et un premier identifiant de chaîne de services, le premier identifiant de chaîne de services indiquant que le premier flux de services est un flux de services de la première chaîne de services, et le premier horodatage indiquant une heure de création du premier flux de services ; la détermination, à partir de la première période de temps effective et de la deuxième période de temps effective, d'une période de temps qui comprend le premier horodatage (108) ;

    la détermination, selon une entrée de flux correspondant à la période de temps déterminée, d'un nœud de service pour recevoir le premier paquet (109) ; et

    l'envoi du premier paquet au nœud de service déterminé (110).


     
    2. Procédé selon la revendication 1, comprenant en outre, après la réception d'une deuxième entrée de flux (106) :

    la détermination si une heure de début de la deuxième période de temps effective précède une heure de fin de la première période de temps effective ; et

    si l'heure de début de la deuxième période de temps effective précède l'heure de fin de la première période de temps effective, le changement de l'heure de fin de la première période de temps effective en heure de début de la deuxième période de temps effective.


     
    3. Procédé selon la revendication 1 ou 2, dans lequel la période de temps qui est déterminée à partir de la première période de temps effective et de la deuxième période de temps effective et qui comprend le premier horodatage est la première période de temps effective ; et
    la détermination, selon la première entrée de flux, d'un nœud de service pour recevoir le premier paquet (109) comprend :

    la détermination d'au moins deux nœuds de service selon un nœud d'entrée du premier paquet et la première relation de mappage de voie d'acheminement (1091) ;

    la génération d'un facteur d'équilibrage de charge selon le premier horodatage (1092) ; et

    la détermination, à partir des au moins deux nœuds de service selon les premières proportions de charge respectives des au moins deux nœuds de service et le facteur d'équilibrage de charge, d'un premier nœud de service pour recevoir le premier paquet (1093).


     
    4. Procédé selon la revendication 1 ou 2, dans lequel le premier paquet comprend en outre un facteur d'équilibrage de charge, et la période de temps qui est déterminée à partir de la première période de temps effective et de la deuxième période de temps effective et qui comprend le premier horodatage est la première période de temps effective ; et la détermination, selon la première entrée de flux, d'un nœud de service pour recevoir le premier paquet (109) comprend :

    la détermination d'au moins deux nœuds de service selon un nœud d'entrée du premier paquet et la première relation de mappage de voie d'acheminement (1091) ; et

    la détermination, à partir des au moins deux nœuds de service selon les premières proportions de charge respectives des au moins deux nœuds de service et le facteur d'équilibrage de charge, d'un premier nœud de service pour recevoir le premier paquet (1093).


     
    5. Procédé selon la revendication 3 ou 4, dans lequel le premier paquet est un paquet de tête du premier flux de services, et après la détermination, à partir des au moins deux nœuds de service, d'un premier nœud de service pour recevoir le premier paquet (1093), le procédé comprend en outre :

    la réception d'un deuxième paquet du premier flux de services (111) ; et

    l'envoi du deuxième paquet au premier nœud de service (112).


     
    6. Procédé selon la revendication 5, comprenant en outre, après la détermination, à partir des au moins deux nœuds de service, d'un premier nœud de service pour recevoir le premier paquet (1093) :

    le démarrage d'un temporisateur d'expiration (113) ; et

    comprenant en outre, avant l'envoi du deuxième paquet au premier nœud de service (112) :
    la détermination que le temporisateur d'expiration n'expire pas (114).


     
    7. Procédé de détermination de voie d'acheminement, comprenant :

    la réception d'un premier paquet d'un premier flux de services (201) ;

    la détermination, selon un champ de caractéristiques compris dans le premier paquet, que le premier flux de services est un flux de services d'une première chaîne de services (202) ;

    l'ajout d'un premier horodatage du premier paquet et d'un premier identifiant de chaîne de services au premier paquet, le premier identifiant de chaîne de services indiquant que le premier flux de services est un flux de services de la première chaîne de services,

    et le premier horodatage indiquant une heure de création du premier flux de services (203) ; et

    l'envoi du premier paquet à un commutateur (204) adapté pour effectuer les étapes de procédé selon la revendication 1.


     
    8. Procédé selon la revendication 7, comprenant en outre, avant l'envoi du premier paquet à un commutateur (203) :
    la détermination d'un facteur d'équilibrage de charge du premier paquet, et l'ajout du facteur d'équilibrage de charge au premier paquet.
     
    9. Procédé de détermination de voie d'acheminement, comprenant :

    l'obtention d'une première charge actuelle de chaque nœud de service existant à un premier instant (101) ;

    la génération d'une première entrée de flux, la première entrée de flux comprenant une première relation de mappage de voie d'acheminement d'un flux de services d'une première chaîne de services, une première proportion de charge de chaque nœud de service et une première période de temps effective de la première entrée de flux, et la première proportion de charge étant générée selon la première charge actuelle de chaque nœud de service (101) ;

    l'envoi de la première entrée de flux à un commutateur (102) ;

    l'obtention d'une deuxième charge actuelle de chaque nœud de service existant à un deuxième instant (104) ;

    la génération d'une deuxième entrée de flux, la deuxième entrée de flux comprenant une deuxième relation de mappage de voie d'acheminement du flux de services de la première chaîne de services, une deuxième proportion de charge de chaque nœud de service, et une deuxième période de temps effective de la deuxième entrée de flux, la deuxième proportion de charge étant générée selon la deuxième charge actuelle de chaque nœud de service, et la deuxième période de temps effective étant différente de la première période de temps effective (104) ; et

    l'envoi de la deuxième entrée de flux au commutateur adapté pour effectuer l'étape de procédé selon la revendication 1.


     
    10. Appareil de détermination de voie d'acheminement (300), configuré pour effectuer le procédé selon l'une quelconque des revendications 1 à 6.
     
    11. Appareil de détermination de voie d'acheminement (400), configuré pour effectuer le procédé selon la revendication 7 ou 8.
     
    12. Appareil de détermination de voie d'acheminement (500), comprenant :

    un premier module d'obtention (501), configuré pour obtenir une première charge actuelle de chaque nœud de service existant à un premier instant ;

    un premier module de génération (502), configuré pour générer une première entrée de flux, la première entrée de flux comprenant une première relation de mappage de voie d'acheminement d'un flux de services d'une première chaîne de services, une première proportion de charge de chaque nœud de service et une première période de temps effective de la première entrée de flux, et la première proportion de charge étant générée selon la première charge actuelle de chaque nœud de service ;

    un module d'envoi (503), configuré pour envoyer la première entrée de flux à un commutateur ;

    un deuxième module d'obtention (504), configuré pour obtenir une deuxième charge actuelle de chaque nœud de service existant à un deuxième instant ; et

    un deuxième module de génération (505), configuré pour générer une deuxième entrée de flux, la deuxième entrée de flux comprenant une deuxième relation de mappage de voie d'acheminement du flux de services de la première chaîne de services, une deuxième proportion de charge de chaque nœud de service et une deuxième période de temps effective de la deuxième entrée de flux, la deuxième proportion de charge étant générée selon la deuxième charge actuelle de chaque nœud de service, et la deuxième période de temps effective étant différente de la première période de temps effective, dans lequel

    le module d'envoi est en outre configuré pour envoyer la deuxième entrée de flux au commutateur adapté pour effectuer le procédé selon la revendication 1.


     
    13. Système de détermination de voie d'acheminement, comprenant :

    un contrôleur (701), configuré pour : obtenir une première charge actuelle de chaque nœud de service existant à un premier instant ; générer une première entrée de flux, la première entrée de flux comprenant une première relation de mappage de voie d'acheminement d'un flux de services d'une première chaîne de services, une première proportion de charge de chaque nœud de service et une première période de temps effective de la première entrée de flux, et la première proportion de charge étant générée selon la première charge actuelle de chaque nœud de service ; envoyer la première entrée de flux à un commutateur ; obtenir la deuxième charge actuelle de chaque nœud de service existant à un deuxième instant ; générer une deuxième entrée de flux, la deuxième entrée de flux comprenant une deuxième relation de mappage de voie d'acheminement d'un paquet du premier type de flux de services, une deuxième proportion de charge de chaque nœud de service et une deuxième période de temps effective de la deuxième entrée de flux, et la deuxième proportion de charge étant générée selon la deuxième charge actuelle de chaque nœud de service ; et envoyer la deuxième entrée de flux au commutateur, la deuxième période de temps effective étant différente de la première période de temps effective ;

    un classificateur (702), configuré pour : recevoir un premier paquet d'un premier flux de services ; déterminer, selon un champ de caractéristiques compris dans le premier paquet, que le premier flux de services est un flux de services de la première chaîne de services ; ajouter un premier horodatage du premier paquet et un premier identifiant de chaîne de services au premier paquet, le premier identifiant de chaîne de services indiquant que le premier flux de services est un flux de services de la première chaîne de services, et le premier horodatage indiquant une heure de création du premier flux de services ; et envoyer le premier paquet au commutateur ; et

    le commutateur (703), configuré pour : recevoir la première entrée de flux ; recevoir la deuxième entrée de flux ; recevoir le premier paquet du premier flux de services ; déterminer, à partir de la première période de temps effective et de la deuxième période de temps effective, une période de temps qui comprend le premier horodatage ;

    déterminer, selon une entrée de flux correspondant à la période de temps déterminée, un nœud de service pour recevoir le premier paquet ; et envoyer le premier paquet au nœud de service déterminé.


     
    14. Système selon la revendication 13, dans lequel le commutateur est en outre configuré pour : après la réception de la deuxième entrée de flux,
    déterminer si une heure de début de la deuxième période de temps effective précède une heure de fin de la première période de temps effective ; et
    si l'heure de début de la deuxième période de temps effective précède l'heure de fin de la première période de temps effective, changer l'heure de fin de la première période de temps effective en heure de début de la deuxième période de temps effective.
     
    15. Système selon la revendication 13 ou 14, dans lequel le premier paquet est un paquet de tête du premier flux de services ; et
    le commutateur est en outre configuré pour : recevoir un deuxième paquet du premier flux de services ; et envoyer le deuxième paquet à un premier nœud de service.
     




    Drawing























    Cited references

    REFERENCES CITED IN THE DESCRIPTION



    This list of references cited by the applicant is for the reader's convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

    Patent documents cited in the description