(19)
(11)EP 2 887 720 B1

(12)EUROPEAN PATENT SPECIFICATION

(45)Mention of the grant of the patent:
11.12.2019 Bulletin 2019/50

(21)Application number: 13831485.1

(22)Date of filing:  19.08.2013
(51)Int. Cl.: 
H04W 16/10  (2009.01)
H04W 88/12  (2009.01)
(86)International application number:
PCT/CN2013/081743
(87)International publication number:
WO 2014/029308 (27.02.2014 Gazette  2014/09)

(54)

RESOURCE ALLOCATION METHOD AND DEVICE

RESSOURCENZUWEISUNGSVERFAHREN UND -VORRICHTUNG

PROCÉDÉ ET DISPOSITIF D'ATTRIBUTION DE RESSOURCES


(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

(30)Priority: 20.08.2012 CN 201210295837

(43)Date of publication of application:
24.06.2015 Bulletin 2015/26

(73)Proprietor: ZTE Corporation
Shenzhen, Guangdong 518057 (CN)

(72)Inventor:
  • YAN, Juming
    Shenzhen Guangdong 518057 (CN)

(74)Representative: Chimini, Francesco 
Jacobacci & Partners S.p.A. Piazza della Vittoria 11
25122 Brescia
25122 Brescia (IT)


(56)References cited: : 
WO-A1-2012/097875
CN-A- 101 114 927
US-A1- 2004 009 773
US-A1- 2012 198 465
WO-A2-2011/044396
CN-A- 101 184 114
US-A1- 2009 228 890
  
      
    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, and in particular to a resource allocation method and device.

    BACKGROUND ART



    [0002] In a mobile communication system, a radio network controller (RNC for short) divides service processing single boards into two categories according to the difference in processing protocols and functions:
    1. 1) a control plane processing board (CP for short), mainly used for processing transmission signalling and a wireless air interface control plane protocol; and
    2. 2) a user plane processing board (UP for short), mainly used for processing a wireless user plane protocol.


    [0003] The CP and UP may further be divided into two categories respectively, one category being common CP and common UP, mainly used for processing protocols of respective common parts, for example: protocols of a cell, a common channel, etc.; and the other category being dedicated CP and dedicated UP, mainly used for processing functions relevant to specific dedicated user equipment (UE for short). The backup mode of the common CP and common UP is 1+1 backup, whereas that of the dedicated CP and dedicated UP is a resource pool manner, i.e. there are a dedicated CP resource pool and a dedicated UP resource pool.

    [0004] In practical network operations, both the traditional dedicated CP resource pool and dedicated UP resource pool have a set of static configuration proportion, solidifying the number and proportion of the CP resource pool and the UP resource pool. With the decrease of the price of smart terminals and the popularization of Android system and smart mobile phones, there are an increasing number of smart terminals in a network. However, due to the requirement of power saving, a smart terminal may automatically initiate the change of a state, and its behaviour is not limited by the control of the RNC, thereby increasing the load of control plane signalling processing. From the analysis in terms of the service mode, a smart mobile phone user may frequently receive and send a mail, perform an instant messaging (IM for short) message call, etc. The time interval between receiving and sending may be several seconds. During the interval, in order to save power, the smart mobile phone generally actively releases a signalling connection and enters an idle mode. In this way, much connection/release will be caused, which occupies network resources, results in network congestion, and increases the load of the control plane (CP) and the user plane (UP) of the controller.

    [0005] Applying a solidification method to allocation of nodes in a resource pool causes that resources cannot be utilized reasonably, which results in the waste of resource, and is also hardly to meet the complicated and variable service distribution and signalling storm in the existing wireless networks gradually. US 2009228890 provides another solution, but the above problem, however, still remains unsolved. Document US 2012/0198465 A1 relates to a system and method for massively multi-core computing, and document WO 2012/097875 A1 refers to a method in a mobility management control node. According to WO 2011/044396 A2 nodes are to be improved by supporting subscriber-based activities of a switch within the nodes.

    [0006] Aiming at the problem above, no effective solution has been presented.

    SUMMARY



    [0007] The invention is defined by independent claims 1 and 8. In the following, parts of the description and drawings referring to embodiments which are not covered by the claims are not presented as embodiments of the invention, but as examples useful for understanding the invention. An embodiment of the present invention provides a resource allocation method and device, so as to at least solve the technical problem of the waste of resources caused by the static configuration of resources in a resource pool in the related art.

    [0008] According to one embodiment of the present invention, a resource allocation method as according to claim 1 is provided.

    [0009] In an example embodiment, according to the monitoring result, adjusting the attribute of some or all nodes in the plurality of nodes includes: when it is detected that a total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and a total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of UP to be CP; and/or when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of CP to be UP.

    [0010] In an example embodiment, a ratio value of the number of users currently accessing a system to a maximum number of users capable of being supported by each node with the attribute of CP is taken as the current total load of the at least one node with the attribute of CP, wherein the system is a mobile communication system; and/or the current total load of the at least one node with the attribute of UP is determined according to a current central processing unit (CPU) load and media plane traffic.

    [0011] In an example embodiment, the one or more nodes of which the attribute is adjusted are one or more nodes with a minimum current load.

    [0012] In an example embodiment, before adjusting the attribute of one or more nodes in the plurality of nodes, the method further includes: judging whether one or more remaining nodes with the same attribute as the adjusted one or more nodes are capable of bearing, after the attribute of the one or more nodes is adjusted, a service load corresponding to the attribute.

    [0013] In an example embodiment, adjusting the attribute of the one or more nodes in the plurality of nodes includes: sending an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and if an adjustment completion indication message returned from the node is not received within a predetermined time, recovering the attribute of the node to the attribute before adjustment.

    [0014] In an example embodiment, monitoring the plurality of nodes of the shared resource pool includes: periodically reporting, by each node in the shared resource pool, a current load of the node and a processing capability of the node, wherein the processing capability refers to the CP load or UP load which can be supported by the nodes; or periodically detecting, by a resource configuration management module, the current load and the processing capability of each node in the shared resource pool.

    [0015] In an example embodiment, the CP is dedicated CP, and the UP is dedicated UP.

    [0016] According to another embodiment of the present invention, a resource allocation device as according to claim 8 is provided.

    [0017] In an example embodiment, the adjustment unit includes: a first adjustment module configured to, when it is detected that a total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and a total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of UP to be CP; and/or a second adjustment module configured to, when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of CP to be UP.

    [0018] In an example embodiment, the adjustment unit further includes: a sending module configured to send an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and a recovery module configured to, if an adjustment completion indication message returned from the node is not received within a predetermined time, recover the attribute of the node to the attribute before adjustment.

    [0019] In the embodiments of the present invention, the CP node and the UP node are allocated by means of a shared resource pool, which composes of a dedicated CP resource pool and dedicated UP resource pool as according to claims 1 and 8, such that dynamic switching may be implemented between the CP node and the UP node. The purpose of effectively utilizing resources may be achieved as long as the resources in the resource pool are monitored in real time and switching between node attributes is implemented according to the monitoring result. The method above solves the technical problem of the waste of resources caused by the static configuration of resources in a resource pool in the related art, and achieves the technical effect of effectively utilizing resources and improving the service processing capability.

    BRIEF DESCRIPTION OF THE DRAWINGS



    [0020] Drawings, provided for further understanding of the present invention and forming a part of the specification, are used to explain the present invention together with embodiments of the present invention rather than to limit the present invention. In the accompanying drawings:

    Fig. 1 is an example flowchart of a resource allocation method according to an embodiment of the present invention;

    Fig. 2 is another example flowchart of a resource allocation method according to an embodiment of the present invention;

    Fig. 3 is an example structural block diagram of a resource allocation device according to an embodiment of the present invention;

    Fig. 4 is another example structural block diagram of a resource allocation device according to an embodiment of the present invention;

    Fig. 5 is yet another example structural block diagram of a resource allocation device according to an embodiment of the present invention;

    Fig. 6 is yet another example flowchart of a resource allocation method according to an embodiment of the present invention;

    Fig. 7 is yet another example flowchart of a resource allocation method according to an embodiment of the present invention;

    Fig. 8 is an example schematic structural diagram of a shared resource pool system according to an embodiment of the present invention;

    Fig. 9 is an example schematic diagram of initial allocation of a shared resource pool according to an embodiment of the present invention;

    Fig. 10 is a flowchart of CP-to-UP attribute switching processing according to an embodiment of the present invention; and

    Fig. 11 is a flowchart of UP-to-CP attribute switching processing according to an embodiment of the present invention.


    DETAILED DESCRIPTION OF THE EMBODIMENTS



    [0021] The present invention is described below with reference to the accompanying drawings and embodiments in detail. Note that, the embodiments of the present invention and the features of the embodiments can be combined with each other if there is no conflict.

    [0022] An example embodiment of the present invention provides a resource allocation method. As shown in Fig. 1, the method includes the steps of:

    step S102: monitoring a plurality of nodes of a shared resource pool, wherein the plurality of nodes include at least one node with the attribute of CP and at least one node with the attribute of UP; and

    step S104: according to the monitoring result, adjusting the attribute of one or more nodes in the plurality of nodes.



    [0023] In the example implementation above, the CP node and the UP node are allocated by means of a shared resource pool, such that dynamic switching may be implemented between the CP node and the UP. The purpose of effectively utilizing resources can be achieved as long as the resources in the resource pool are monitored in real time and switching between node attributes is implemented according to the monitoring result. The method above solves the technical problem of the waste of resources caused by the static configuration of resources in a resource pool in the related art, and achieves the technical effect of effectively utilizing resources and improving the service processing capability.

    [0024] As regards the adjustment of the attribute of one or more nodes in the plurality of nodes performed according to the monitoring result, there are mainly two cases, one being to adjust the attribute of the node with the attribute of CP to be UP, and the other being to adjust the attribute of the node with the attribute of UP to be CP. Such adjustment above is mainly to implement load balancing, i.e. corresponding adjustment is performed mainly according to the processing load of the nodes with each attribute. In an example implementation, according to the monitoring result, adjusting the attribute of some or all nodes in the plurality of nodes includes:
    1. 1) when it is detected that the total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and the total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of UP to be CP; and/or
    2. 2) when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of CP to be UP.


    [0025] By means of performing adjustment according to respective loads, load balancing of nodes in a resource pool can be effectively achieved.

    [0026] The determination of the load of CP and UP may apply different bases. For example, as regards the at least one node with the attribute of CP, the ratio value of the number of users currently accessing the system to the maximum number of users capable of being supported by each node with the attribute of CP may be taken as the current total load of the at least one node with the attribute of CP; and as regards the at least one node with the attribute of UP, the current total load of the at least one node with the attribute of UP may be determined according to the current CPU load and media plane traffic.

    [0027] In order to ensure the minimum change to nodes, when it is determined that the attribute of a node needs to be adjusted, a processing method according to an example embodiment is to choose the node with the minimum current load to be adjusted. For example, when the load of CP is excessively great and one or more UP nodes need to be converted into one or more CP nodes, the one or more nodes with the minimum load in the UP nodes may be chosen to be converted. That is to say, the one or more nodes of which the attribute is adjusted are one or more nodes with the minimum current load.

    [0028] In consideration of the number of nodes corresponding to the adjusted nodes possibly being hardly to bear the current service of the node after the node with one attribute is adjusted to be the other attribute, such adjustment is obviously unreasonable. In order to avoid the occurrence of such situation, the following judgement may be performed before adjustment: judging whether one or more remaining nodes with the same attribute as the adjusted one or more nodes are capable of bearing, if (after) the attribute of the one or more nodes is adjusted, a service load corresponding to the attribute.

    [0029] In practical applications, adjustment abnormality may occur; therefore, a fallback operation may be added to the adjustment process. For example, one CP node may need to be adjusted to be a UP node; however, in the adjustment process, the node is not adjusted successfully, and thus an abnormality may occur. Therefore, a timer may be provided, and if a message of successful adjustment returned from the node is not received within a predetermined time, the attribute of the node is still set as CP. In one example implementation, as shown in Fig. 2, adjusting the attribute of one or more nodes in the plurality of nodes includes the steps of:

    step S202: sending an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and

    step S204: if an adjustment completion indication message returned from the node is not received within a predetermined time, recovering the attribute of the node to the attribute before adjustment.



    [0030] The plurality of nodes in the shared resource pool may be monitored according to information actively reported by a node, or according to information actively checked by a resource configuration management module. In an example embodiment, the plurality of nodes in the shared resource pool may be monitored by but not limited to one of the following means:
    1. 1) periodically reporting, by each node in the shared resource pool, a current load of the node and a processing capability of the node; or
    2. 2) periodically detecting, by a resource configuration management module, the current load and the processing capability of each node in the shared resource pool.


    [0031] In various example implementations above, the CP may be a dedicated CP, and the UP may be a dedicated UP.

    [0032] A resource allocation device is further provided in this embodiment. The device is used for implementing the embodiments and example implementations above, and what has been described will not be stated redundantly. As will be used below, the term "unit" or "module" may be a combination of software and/or hardware implementing a predetermined function. Although the device described in the embodiments below is preferably implemented by means of software, the implementation of hardware or a combination of software and hardware may also be possible and conceivable. Fig. 3 is an example structural block diagram of a resource allocation device according to an embodiment of the present invention. As shown in Fig. 3, the device includes: a monitoring unit 302 and an adjustment unit 304. The structure will be described below.

    [0033] The monitoring unit 302 is configured to monitor a plurality of nodes of a shared resource pool, wherein the plurality of nodes include at least one node with the attribute of CP and at least one node with the attribute of UP; and
    the adjustment unit 304 is coupled with the monitoring unit 302 and is configured to, according to a monitoring result, adjust the attribute of one or more nodes in the plurality of nodes.

    [0034] In an example implementation, as shown in Fig. 4, the adjustment unit includes: a first adjustment module 402 coupled with the monitoring unit 302 and configured to, when it is detected that the total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and the total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of UP to be CP; and/or a second adjustment module 404 coupled with the monitoring unit 302 and configured to, when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of CP to be UP.

    [0035] In an example implementation, as shown in Fig. 5, the adjustment unit may further include: a sending module 502 configured to send an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and a recovery module 504 coupled with the sending module 502 and configured to, if an adjustment completion indication message returned from the node is not received within a predetermined time, recover the attribute of the node to the attribute before adjustment.

    [0036] In order to solve the problem that the traditional solidification configuration cannot adapt to the requirement of variability of the existing wireless network services, a control plane and a user plane processing board share a resource pool in the embodiments of the present invention on the premise of the CP and UP both using the same kind of physical single boards, so as to increase the resource utilization rate and the whole processing capability of the system, and reduce the number of redundant single boards. Specifically, the dedicated CP resource pool and dedicated UP resource pool which are independent originally are combined to form a big resource pool. Each node in the resource pool is completely identical, and meanwhile, according to the current system load, each node may serve as dedicated CP, and may also serve as dedicated UP.

    [0037] For example, with the normal operation of the system, when the whole UP load is very high, while the CP nodes are capable of bearing additional load, a certain node which serves as a CP function in the resource pool may be dynamically switched to serve as a UP function. Alternatively, when the whole CP load is very high, while the UP nodes are capable of bearing additional load, the attribute of a certain node which serves as a UP function in the resource pool may be switched to serve as a CP function.

    [0038] In particular, as shown in Fig. 6, the resource allocation method includes the steps of:

    step S602: calculating the number of nodes of dedicated CP and the number of nodes of dedicated UP required according to the system capacity, and configuring these nodes as nodes in a shared resource pool (in an example embodiment, the external attribute of each node does not need to be set);

    step S604: performing, by the system, load monitoring on each node in the resource pool periodically, wherein as regards CP and UP, the load may be determined according to different conditions, for example, as regards a node with a CP attribute, the load may be but not limited to be a normalization load of the proportion of the number of current users to the maximum number of users supported by a single node; and as regards a node with a UP attribute, the load may be a normalization load after weighting the CPU load, media plane traffic, etc.;

    step S606: after the system is powered on and controlled, reading the configuration of the resource pool, and in the meantime, allocating an initial attribute to each node, i.e. initially setting the node as a CP attribute or a UP attribute;

    step S608: in the process of system operation, any one node in the resource pool periodically reporting the current load and the total processing capability of this node; and

    step S610: periodically performing load evaluation by the system, and when it is found that there is mismatching between the loads of CP resources and UP resources, i.e. when the current system network operation conditions result in insufficient resources of a certain party, while the other party still has a large amount of surplus, using a predetermined algorithm to choose to dynamically switch the node attribute of a certain node.



    [0039] Two specific embodiments will be combined below for illustration, and in this example implementation, a resource pool having N nodes will be taken as an example for illustration.

    [0040] As shown in Fig. 7, the resource allocation method includes the steps as follows:

    step S702: single boards in the shared resource pool and related resource allocation are configured in background, such that a resource configuration management module in foreground knows how much capacity of a resource pool has been configured.

    Step S704: as shown in Fig. 8, a resource configuration management module is deployed on a master control single board for globally evaluating the load in the resource pool and making a corresponding judgement. In an example embodiment, one resource configuration management proxy may be deployed in each resource pool node for calculating the load of this node, and then reporting the load to the resource configuration management module, and meanwhile, the resource configuration management proxy is also used for receiving a judgement command of the resource configuration management module.

    Step S706: the whole system is powered on, and the resource configuration management module acquires resource pool information. As shown in Fig. 9, according to the traditional configuration, firstly an initial attribute is set for any one node in the resource pool, i.e. this node is with the attribute of CP or the attribute of UP.

    Step S708: each node in the resource pool periodically (e.g. every 2s) reports the processing capability and the current load of this node to the resource configuration management module of the master control single board via the resource configuration proxy.

    Step S710: the resource configuration management module of the master control module periodically performs load evaluation, mainly including:

    1. 1) calculating the total CP processing capability of all nodes which are CP currently and operate normally in the resource pool, and the current total CP load, and
    2. 2) calculating the total UP processing capability of all nodes which are UP currently and operate normally in the resource pool, and the current total UP load.



    [0041] Two specific implementations are illustrated below for describing the method above.

    Implementation 1



    [0042] This implementation describes switching a resource pool single board from a control plane (CP) to a user plane (UP). As shown in Fig. 10, the CP-to-UP attribute switching processing includes the steps of:

    step S 1002: after performing evaluation according to the load, making, by a resource configuration management module, the following judgement according to different scenarios:

    scenario A: if the current total CP load is a light load state compared to the total processing capability (in an example embodiment, a threshold value may be decided by means of configuration), powering off a CP node with the minimum load in the resource pool;

    scenario B: if the current total CP load is a non-light load state compared to the total processing capability but the processing capability has surplus, maintaining this state; and

    scenario C: if the current total UP load is a high load state compared to the total processing capability, the processing capability cannot support the current load, and some CP nodes in the system need to be switched to the UP attribute, performing the processing of:

    S1: judging whether there is a UP node which has been powered off in the resource pool, and if so, directly recovering the UP node which has been powered off to be powered on;

    S2: if there is no UP node which has been powered off, then judging whether there is a CP node which has been powered off in the resource pool, and if so, powering on the CP node which has been powered off, and switching the attribute of the node to UP;

    S3: if there is no CP node which has been powered off in the resource pool, then choosing a CP node with the lowest load from a CP node pool, and meanwhile, calculating whether the total remaining CP processing capability after removing the calculation capability of the CP node is sufficient to support the current CP load, and if so, determining to switch the attribute of the node from CP to UP; and

    S4: if the total remaining CP processing capability after removing the calculation capability of the CP node is insufficient to support the current CP load, maintaining this state;

    step S1004: after the resource configuration management module judges that it is needed to switch the CP node from the CP attribute to UP, sending an attribute switching command to the resource configuration proxy of the CP node needing to be switched;

    step S 1006: after receiving the attribute switching command, sending, by the resource configuration proxy, to the CP service a prenotification message to notify the node that the attribute will be switched to UP soon, and providing a waiting timer, the timer being used for judging whether the switching can be successfully completed within a predetermined time; after having received the message, judging, by the CP service, whether the service volume, service form, etc. on the current CP node allows switching, and if so, starting to perform relevant service migration, etc.; and after service aftermath work is processed completely, returning an attribute switching response message to the resource configuration proxy; and

    step S1008: judging, by the resource configuration proxy, the response message, mainly including:

    1. 1) if the service in the response message acknowledges that the CP node can perform attribute switching, then performing, by the resource configuration proxy, operation software switching, and operating UP software; and then notifying the resource configuration management module to update information and relevant node states; and
    2. 2) if the service in the response message acknowledges that the current CP node cannot perform attribute switching due to some situations, then terminating, by the resource configuration proxy, a final switching action, and then notifying the resource configuration management module to perform state fallback, the attribute of the CP node still being CP.


    Implementation 2



    [0043] This implementation describes switching a resource pool single board from a user plane (UP) to a control plane (CP). As shown in Fig. 11, the UP-to-CP attribute switching processing includes the steps of:

    step S 1102: after performing evaluation according to the load, making, by a resource configuration management module, the following judgement according to different scenarios:

    scenario A: if the current total UP load is a light load state compared to the total processing capability (in an example embodiment, a threshold value may be decided by means of configuration), powering off a UP node with the minimum load in the resource pool;

    scenario B: if the current total UP load is a non-light load state compared to the total processing capability but the processing capability has surplus, maintaining this state; and

    scenario C: if the current total CP load is a high load state compared to the total processing capability, the processing capability cannot support the current load, and some UP nodes in the system need to be switched to the CP attribute, performing the processing of:

    S1: judging whether there is a CP node which has been powered off in the resource pool, and if so, directly recovering the CP node which has been powered off to be powered on;

    S2: if there is no CP node which has been powered off, then judging whether there is a UP node which has been powered off in the resource pool, and if so, powering on the UP node which has been powered off, and switching the attribute of the node to CP;

    S3: if there is no UP node which has been powered off in the resource pool, then choosing a UP node with the lowest load from a UP node pool, and meanwhile, calculating whether the total remaining UP processing capability after removing the calculation capability of the UP node is sufficient to support the current UP load, and if so, determining to switch the attribute of the node from UP to CP; and

    S4: if the total remaining UP processing capability after removing the calculation capability of the UP node is insufficient to support the current UP load, maintaining this state;

    step S1104: after the resource configuration management module judges that it is needed to switch the UP node from the UP attribute to CP, sending an attribute switching command to the resource configuration proxy of the UP node needing to be switched;

    step S1106: after receiving the attribute switching command, sending, by the resource configuration proxy, to the UP service a prenotification message to notify the node that the attribute will be switched to CP soon, and providing a waiting timer, the timer being used for judging whether the switching can be successfully completed within a predetermined time; after having received the message, judging, by the UP service, whether the service volume, service form, etc. on the current UP node allows switching, and if so, starting to perform relevant service migration, etc.; and after service aftermath work is processed completely, returning an attribute switching response message to the resource configuration proxy; and

    step S1108: judging, by the resource configuration proxy, the response message, mainly including:

    1. 1) if the service in the response message acknowledges that the UP node can perform attribute switching, then performing, by the resource configuration proxy, operation software switching, and operating CP software; and then notifying the resource configuration management module to update information and relevant node states; and
    2. 2) if the service in the response message acknowledges that the current UP node cannot perform attribute switching due to some situations, then terminating, by the resource configuration proxy, a final switching action, and then notifying the resource configuration management module to perform state fallback, the attribute of the UP node still being UP.



    [0044] From the description above, it can be seen that the embodiments of the present invention achieve the technical effects as follows: the CP node and the UP node are allocated by means of a shared resource pool, such that dynamic switching may be implemented between the CP node and the UP. The purpose of effectively utilizing resources can be achieved as long as the resources in the resource pool are monitored in real time and switching between node attributes is implemented according to the monitoring result. The method above solves the technical problem of the waste of resources caused by the static configuration of resources in a resource pool in the related art, and achieves the technical effect of effectively utilizing resources and improving the service processing capability.

    [0045] Obviously, those skilled in the art should know that each of the mentioned modules or steps of the present invention can be realized by universal computing devices; the modules or steps can be focused on single computing device, or distributed on the network formed by multiple computing devices; selectively, they can be realized by the program codes which can be executed by the computing device; thereby, the modules or steps can be stored in the storage device and executed by the computing device.


    Claims

    1. A resource allocation method, characterized by comprising:

    monitoring a plurality of nodes of a shared resource pool, wherein the plurality of nodes comprise at least one node with an attribute of control plane processing board, CP, and at least one node with an attribute of user plane processing board, UP (102); and

    according to a monitoring result, adjusting the attribute of one or more nodes in the plurality of nodes (104);

    wherein, the shared resource pool composes of a dedicated CP resource pool and dedicated UP resource pool;

    the CP is dedicated CP, and the UP is dedicated UP, wherein the dedicated CP and dedicated UP, are used for processing functions relevant to specific dedicated user equipment, and the backup mode of the dedicated CP or dedicated UP is a resource pool manner.


     
    2. The method according to claim 1, characterized in that according to the monitoring result, adjusting the attribute of some or all nodes in the plurality of nodes comprises:

    when it is detected that a total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and a total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of UP to be CP; and/or

    when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjusting the attribute of one or more nodes with the current attribute of CP to be UP.


     
    3. The method according to claim 2, characterized in that

    a ratio value of the number of users currently accessing a system to a maximum number of users capable of being supported by each node with the attribute of CP is taken as the current total load of the at least one node with the attribute of CP, wherein the system is a mobile communication system; and/or

    the current total load of the at least one node with the attribute of UP is determined according to a current central processing unit, CPU, load and media plane traffic.


     
    4. The method according to claim 1, characterized in that the one or more nodes of which the attribute is adjusted are one or more nodes with a minimum current load.
     
    5. The method according to claim 1, characterized in that before adjusting the attribute of the one or more nodes in the plurality of nodes, the method further comprises:
    judging whether one or more remaining nodes with the same attribute as the adjusted one or more nodes are capable of bearing, after the attribute of the one or more nodes is adjusted, a service load corresponding to the attribute.
     
    6. The method according to any one of claims 1 to 5, characterized in that adjusting the attribute of the one or more nodes in the plurality of nodes comprises:

    sending an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and

    if an adjustment completion indication message returned from the node is not received within a predetermined time, recovering the attribute of the node to the attribute before adjustment.


     
    7. The method according to any one of claims 1 to 5, characterized in that monitoring the plurality of nodes of the shared resource pool comprises:

    periodically reporting, by each node in the shared resource pool, a current load of the node and a processing capability of the node, wherein the processing capability refers to the CP load or UP load which can be supported by the nodes; or

    periodically detecting, by a resource configuration management module, the current load and the processing capability of each node in the shared resource pool.


     
    8. A resource allocation device, characterized by comprising:

    a monitoring unit (302) configured to monitor a plurality of nodes of a shared resource pool, wherein the plurality of nodes comprise at least one node with an attribute of control plane processing board, CP, and at least one node with an attribute of user plane processing board, UP; and

    an adjustment unit (304) configured to, according to a monitoring result, adjust the attribute of one or more nodes in the plurality of nodes;

    wherein, the shared resource pool composes of a dedicated CP resource pool and dedicated UP resource pool;

    the CP is dedicated CP, and the UP is dedicated UP, wherein the dedicated CP and dedicated UP, are used for processing functions relevant to specific dedicated user equipment, and the backup mode of the dedicated CP or dedicated UP is a resource pool manner.


     
    9. The device according to claim 8, characterized in that the adjustment unit (304) comprises:

    a first adjustment module (402) configured to, when it is detected that a total load of the at least one node with the current attribute of CP is greater than a first predetermined threshold value, and a total load of the at least one node with the current attribute of UP is smaller than a second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of UP to be CP; and/or

    a second adjustment module (404) configured to, when it is detected that the total load of the at least one node with the current attribute of UP is greater than a third predetermined threshold value, and the total load of the at least one node with the current attribute of CP is smaller than the second predetermined threshold value, adjust the attribute of one or more nodes with the current attribute of CP to be UP.


     
    10. The device according to claim 9, characterized in that

    a ratio value of the number of users currently accessing a system to a maximum number of users capable of being supported by each node with the attribute of CP is taken as the current total load of the at least one node with the attribute of CP, wherein the system is a mobile communication system; and/or

    the current total load of the at least one node with the attribute of UP is determined according to a current central processing unit, CPU, load and media plane traffic.


     
    11. The device according to claim 8, characterized in that the one or more nodes of which the attribute is adjusted are one or more nodes with a minimum current load.
     
    12. The device according to claim 8, characterized in that the device is further configured to, before the adjustment unit adjusts the attribute of the one or more nodes in the plurality of nodes, judge whether one or more remaining nodes with the same attribute as the adjusted one or more nodes are capable of bearing, after the attribute of the one or more nodes is adjusted, a service load corresponding to the attribute.
     
    13. The device according to any one of claims 8 to 12, characterized in that the adjustment unit (304) further comprises:

    a sending module (502) configured to send an attribute adjustment instruction to each node to be adjusted, wherein the attribute adjustment instruction is used for instructing the node to change its current attribute; and

    a recovery module (504) configured to, if an adjustment completion indication message returned from the node is not received within a predetermined time, recover the attribute of the node to the attribute before adjustment.


     
    14. The device according to any one of claims 8 to 12, characterized in that the monitoring unit (302) is configured to monitor the plurality of nodes of the shared resource pool in the following manner:

    periodically reporting, by each node in the shared resource pool, a current load of the node and a processing capability of the node, wherein the processing capability refers to the CP load or UP load which can be supported by the nodes; or

    periodically detecting, by a resource configuration management module, the current load and the processing capability of each node in the shared resource pool.


     


    Ansprüche

    1. Ressourcenzuweisungsverfahren, dadurch gekennzeichnet, dass es umfasst:

    Überwachen einer Vielzahl von Knoten eines gemeinsam genutzten Ressourcenpools, wobei die Vielzahl von Knoten mindestens einen Knoten mit einem Attribut einer Steuer- bzw. Regelebenen-Verarbeitungsplatine, CP, und mindestens einen Knoten mit einem Attribut einer Benutzerebenen-Verarbeitungsplatine, UP, (102) umfasst; und

    Anpassen des Attributs eines Knotens oder mehrerer Knoten in der Vielzahl von Knoten (104) gemäß einem Überwachungsergebnis;

    wobei sich der gemeinsam genutzte Ressourcenpool aus einem zweckgebundenen CP-Ressourcenpool und einem zweckgebundenen UP-Ressourcenpool zusammensetzt;

    wobei die CP eine zweckgebundene CP ist, und die UP eine zweckgebundene UP ist, wobei die zweckgebundene CP und die zweckgebundene UP für Verarbeitungsfunktionen verwendet werden, die für bestimmte zweckgebundene Endgeräte relevant sind, und der Backup-Modus der zweckgebundenen CP oder der zweckgebundenen UP ein Ressourcenpool ist.


     
    2. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass das Anpassen des Attributs einiger oder aller Knoten in der Vielzahl von Knoten gemäß dem Überwachungsergebnis umfasst:

    Anpassen des Attributs eines Knotens oder mehrerer Knoten mit dem aktuellen UP-Attribut, sodass es CP ist, wenn erfasst wird, dass eine Gesamtlast des mindestens einen Knotens mit dem aktuellen CP-Attribut größer ist als ein erster vorbestimmter Schwellenwert, und eine Gesamtlast des mindestens einen Knotens mit dem aktuellen UP-Attribut kleiner ist als ein zweiter vorbestimmter Schwellenwert; und/oder

    Anpassen des Attributs eines Knotens oder mehrerer Knoten mit dem aktuellen CP-Attribut, sodass es UP ist, wenn erfasst wird, dass die Gesamtlast des mindestens einen Knotens mit dem aktuellen UP-Attribut größer ist als ein dritter vorbestimmter Schwellenwert, und die Gesamtlast des mindestens einen Knotens mit dem aktuellen CP-Attribut kleiner ist als der zweite vorbestimmte Schwellenwert.


     
    3. Verfahren nach Anspruch 2, dadurch gekennzeichnet,
    dass ein Verhältniswert der Anzahl der Benutzer, die aktuell auf ein System zugreifen, zu einer maximalen Anzahl der Benutzer, die von jedem Knoten mit dem CP-Attribut unterstützt werden können, als die aktuelle Gesamtlast des mindestens einen Knotens mit dem CP-Attribut angenommen wird, wobei das System ein mobiles Kommunikationssystem ist; und/oder
    die aktuelle Gesamtlast des mindestens einen Knotens mit dem UP-Attribut gemäß einer aktuellen Zentraleinheit, CPU, Last und dem Medienebenenverkehr bestimmt wird.
     
    4. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass der eine oder die mehreren Knoten, dessen/deren Attribut angepasst wird, ein oder mehrere Knoten mit einer minimalen aktuellen Last ist/sind.
     
    5. Verfahren nach Anspruch 1, dadurch gekennzeichnet, dass, vor dem Anpassen des Attributs des einen Knotens oder der mehreren Knoten in der Vielzahl von Knoten, das Verfahren ferner umfasst:
    Beurteilen, ob ein oder mehrere verbleibende/r Knoten mit dem gleichen Attribut wie der angepasste eine oder die angepassten mehreren Knoten in der Lage sind, nachdem das Attribut des einen Knotens oder der mehreren Knoten angepasst ist, eine dem Attribut entsprechende Dienstbelastung zu tragen.
     
    6. Verfahren nach einem beliebigen der Ansprüche 1 bis 5, dadurch gekennzeichnet, dass das Anpassen des Attributs des einen Knotens oder der mehreren Knoten in der Vielzahl von Knoten umfasst:

    Senden einer Attributanpassungsanweisung an jeden anzupassenden Knoten, wobei die Attributanpassungsanweisung verwendet wird, um den Knoten anzuweisen, sein aktuelles Attribut zu ändern; und

    Wiederherstellen des Attributs des Knotens zu dem Attribut vor der Anpassung, wenn eine vom Knoten zurückgegebene Meldung zur Beendigung der Anpassung nicht innerhalb einer vorbestimmten Zeit empfangen wird.


     
    7. Verfahren nach einem beliebigen der Ansprüche 1 bis 5, dadurch gekennzeichnet, dass das Überwachen der Vielzahl von Knoten des gemeinsam genutzten Ressourcenpools umfasst:

    regelmäßiges Berichten einer aktuellen Last des Knotens und einer Verarbeitungsleistung des Knotens von jedem Knoten in dem gemeinsam genutzten Ressourcenpool, wobei sich die Verarbeitungsleistung auf die CP-Last oder UP-Last bezieht, die von den Knoten unterstützt werden kann; oder

    regelmäßiges Erfassen der aktuellen Last und der Verarbeitungsleistung jedes Knotens in dem gemeinsam genutzten Ressourcenpool von einem Ressourcenkonfigurations- Verwaltungsmodul.


     
    8. Ressourcenzuweisungsvorrichtung, dadurch gekennzeichnet, dass sie umfasst:

    eine Überwachungseinheit (302), die ausgestaltet ist, eine Vielzahl von Knoten eines gemeinsam genutzten Ressourcenpools zu überwachen, wobei die Vielzahl von Knoten mindestens einen Knoten mit einem Attribut der Steuer- bzw. Regelebenen-Verarbeitungsplatine, CP, und mindestens einen Knoten mit einem Attribut der Benutzerebenen-Verarbeitungsplatine, UP, umfasst; und

    eine Anpassungseinheit (304), die ausgestaltet ist, das Attribut eines Knotens oder mehrerer Knoten in der Vielzahl von Knoten gemäß einem Überwachungsergebnis anzupassen;

    wobei sich der gemeinsam genutzte Ressourcenpool aus einem zweckgebundenen CP-Ressourcenpool und einem zweckgebundenen UP-Ressourcenpool zusammensetzt;

    wobei die CP eine zweckgebundene CP ist, und die UP eine zweckgebundene UP ist, wobei die zweckgebundene CP und die zweckgebundene UP für Verarbeitungsfunktionen verwendet werden, die für bestimmte zweckgebundene Endgeräte relevant sind, und der Backup-Modus der zweckgebundenen CP oder der zweckgebundenen UP ein Ressourcenpool ist.


     
    9. Vorrichtung nach Anspruch 8, dadurch gekennzeichnet, dass die Anpassungseinheit (304) umfasst:

    ein erstes Anpassungsmodul (402), das ausgestaltet ist, das Attribut eines Knotens oder mehrerer Knoten mit dem aktuellen UP-Attribut anzupassen, sodass es CP ist, wenn erfasst wird, dass eine Gesamtlast des mindestens einen Knotens mit dem aktuellen CP-Attribut größer ist als ein erster vorbestimmter Schwellenwert, und eine Gesamtlast des mindestens einen Knotens mit dem aktuellen UP-Attribut kleiner ist als ein zweiter vorbestimmter Schwellenwert; und/oder

    ein zweites Anpassungsmodul (404), das ausgestaltet ist, das Attribut eines Knotens oder mehrerer Knoten mit dem aktuellen CP-Attribut anzupassen, um UP zu sein, wenn erfasst wird, dass die Gesamtlast des mindestens einen Knotens mit dem aktuellen UP-Attribut größer ist als ein dritter vorbestimmter Schwellenwert, und die Gesamtlast des mindestens einen Knotens mit dem aktuellen CP-Attribut kleiner ist als der zweite vorbestimmte Schwellenwert.


     
    10. Vorrichtung nach Anspruch 9, dadurch gekennzeichnet, dass
    ein Verhältniswert der Anzahl der Benutzer, die aktuell auf ein System zugreifen, zu einer maximalen Anzahl der Benutzer, die von jedem Knoten mit dem CP-Attribut unterstützt werden können, als die aktuelle Gesamtlast des mindestens einen Knotens mit dem CP-Attribut angenommen wird, wobei das System ein mobiles Kommunikationssystem ist; und/oder
    die aktuelle Gesamtlast des mindestens einen Knotens mit dem UP-Attribut gemäß einer aktuellen Zentraleinheit, CPU, Last und dem Medienebenenverkehr bestimmt wird.
     
    11. Vorrichtung nach Anspruch 8, dadurch gekennzeichnet, dass der eine oder die mehreren Knoten, dessen/deren Attribut angepasst wird, ein oder mehrere Knoten mit einer minimalen aktuellen Last ist/sind.
     
    12. Vorrichtung nach Anspruch 8, dadurch gekennzeichnet, dass die Vorrichtung ferner ausgestaltet ist, bevor die Anpassungseinheit das Attribut des einen Knotens oder der mehreren Knoten in der Vielzahl von Knoten anpasst, zu beurteilen, ob ein oder mehrere verbleibende/r Knoten mit dem gleichen Attribut wie der angepasste eine oder die angepassten mehreren Knoten in der Lage sind, nachdem das Attribut des einen Knotens oder der mehreren Knoten angepasst ist, eine dem Attribut entsprechende Dienstbelastung zu tragen.
     
    13. Vorrichtung nach einem beliebigen der Ansprüche 8 bis 12, dadurch gekennzeichnet, dass die Anpassungseinheit (304) ferner umfasst:

    ein Sendemodul (502), das ausgestaltet ist, eine Attributanpassungsanweisung an jeden anzupassenden Knoten zu senden, wobei die Attributanpassungsanweisung verwendet wird, um den Knoten anzuweisen, sein aktuelles Attribut zu ändern; und

    ein Wiederherstellungsmodul (504), das ausgestaltet ist, das Attribut des Knotens zu dem Attribut vor der Anpassung wiederherzustellen, wenn eine vom Knoten zurückgegebene Meldung zur Beendigung der Anpassung nicht innerhalb einer vorbestimmten Zeit empfangen wird.


     
    14. Vorrichtung nach einem beliebigen der Ansprüche 8 bis 12, dadurch gekennzeichnet, dass die Überwachungseinheit (302) ausgestaltet ist, die Vielzahl von Knoten des gemeinsam genutzten Ressourcenpools in folgender Weise zu überwachen:

    regelmäßiges Berichten einer aktuellen Last des Knotens und einer Verarbeitungsleistung des Knotens von jedem Knoten in dem gemeinsam genutzten Ressourcenpool, wobei sich die Verarbeitungsleistung auf die CP-Last oder UP-Last bezieht, die von den Knoten unterstützt werden kann; oder

    regelmäßiges Erfassen der aktuellen Last und der Verarbeitungsleistung jedes Knotens in dem gemeinsam genutzten Ressourcenpool durch ein Ressourcenkonfigurations-Verwaltungsmodul.


     


    Revendications

    1. Procédé d'attribution de ressources, caractérisé en ce qu'il comprend :

    la surveillance d'une pluralité de nœuds d'un groupe de ressources partagées, dans lequel la pluralité de nœuds comprend au moins un nœud avec un attribut de carte de traitement de plan de commande CP et au moins un nœud avec un attribut de carte de traitement de plan utilisateur UP (102) ; et

    conformément à un résultat de surveillance, l'ajustement de l'attribut d'un ou plusieurs nœuds de la pluralité de nœuds (104) ;

    dans lequel le groupe de ressources partagées se compose d'un groupe dédié de ressources CP et d'un groupe dédié de ressources UP ;

    le CP est un CP dédié et l'UP est un UP dédié, dans lequel le CP dédié et l'UP dédié sont utilisés pour des fonctions de traitement se rapportant à l'équipement utilisateur dédié spécifique, et le mode de sauvegarde du CP dédié ou de l'UP dédié est une sorte de groupe de ressources.


     
    2. Procédé selon la revendication 1, caractérisé en ce que, conformément au résultat de surveillance, l'ajustement de l'attribut de certains ou de la totalité des nœuds de la pluralité de nœuds comprend :

    lorsque l'on détecte une charge totale de l'au moins un nœud avec l'attribut actuel de CP est supérieure à une première valeur de seuil prédéterminée et qu'une charge totale de l'au moins un nœud avec l'attribut actuel d'UP est inférieure à une deuxième valeur de seuil prédéterminée, l'ajustement de l'attribut d'un ou plusieurs nœuds avec l'attribut actuel d'UP pour être CP ; et/ou

    lorsque l'on détecte que la charge totale de l'au moins un nœud avec l'attribut actuel d'UP est supérieure à une troisième valeur de seuil prédéterminée et que la charge totale de l'au moins un nœud avec l'attribut actuel de CP est inférieure à la deuxième valeur de seuil prédéterminée, l'ajustement de l'attribut d'un ou plusieurs nœuds avec l'attribut actuel de CP pour être UP.


     
    3. Procédé selon la revendication 2, caractérisé en ce
    qu'une valeur de rapport du nombre d'utilisateurs accédant actuellement à un système à un nombre maximal d'utilisateurs capables d'être supportés par chaque nœud avec l'attribut de CP est prise comme charge totale actuelle de l'au moins un nœud avec l'attribut de CP, dans lequel le système est un système de communication mobile ; et/ou
    que la charge totale actuelle de l'au moins un nœud avec l'attribut d'UP est déterminée conformément à une charge de l'unité de traitement centrale actuelle CPU et d'un trafic de plan média.
     
    4. Procédé selon la revendication 1, caractérisé en ce que les un ou plusieurs nœuds dont l'attribut est ajusté sont un ou plusieurs nœuds avec une charge actuelle minimale.
     
    5. Procédé selon la revendication 1, caractérisé en ce qu'avant d'ajuster l'attribut des un ou plusieurs nœuds de la pluralité de nœuds, le procédé comprend en outre :
    le jugement quant à savoir si un ou plusieurs nœuds restant ayant le même attribut que les un ou plusieurs nœuds ajustés sont capables de supporter, une fois l'attribut des un ou plusieurs nœuds ajusté, une charge de service correspondant à l'attribut.
     
    6. Procédé selon l'une quelconque des revendications 1 à 5, caractérisé en ce que l'ajustement de l'attribut des un ou plusieurs nœuds de la pluralité de nœuds comprend :

    l'envoi d'une instruction d'ajustement d'attribut à chaque nœud à ajuster, dans lequel l'instruction d'ajustement d'attribut est utilisée pour instruire le nœud de changer son attribut actuel ; et

    si un message d'indication d'achèvement de l'ajustement renvoyé par le nœud n'est pas reçu dans une période prédéterminée, le rétablissement de l'attribut du nœud à l'attribut avant l'ajustement.


     
    7. Procédé selon l'une quelconque des revendications 1 à 5, caractérisé en ce que la surveillance de la pluralité de nœuds du groupe de ressources partagées comprend :

    le rapport périodique, par chaque nœud du groupe de ressources partagées, d'une charge actuelle du nœud et d'une capacité de traitement du nœud, dans lequel la capacité de traitement se réfère à la charge CP ou à la charge UP qui peut être supportée par les noeuds ; ou

    la détection périodique, par un module de gestion de configuration de ressources, de la charge actuelle et de la capacité de traitement de chaque nœud dans le groupe de ressources partagées.


     
    8. Dispositif d'attribution de ressources, caractérisé en ce qu'il comprend :

    une unité de surveillance (302) configurée pour surveiller une pluralité de nœuds d'un groupe de ressources partagées, dans lequel la pluralité de nœuds comprend au moins un nœud avec un attribut d'une carte de traitement de plan de commande CP et au moins un nœud avec un attribut de carte de traitement de plan utilisateur UP ; et

    une unité d'ajustement (304) configurée pour, conformément à un résultat de surveillance, ajuster l'attribut d'un ou plusieurs nœuds de la pluralité de noeuds ;

    dans lequel le groupe de ressources partagées se compose d'un groupe dédié de ressources CP et d'un groupe dédié de ressources UP ;

    le CP est un CP dédié et l'UP est un UP dédié, dans lequel le CP dédié et l'UP dédié sont utilisés pour des fonctions de traitement se rapportant à un équipement utilisateur dédié spécifique et le mode de sauvegarde du CP dédié ou de l'UP dédié est une sorte de groupe de ressources.


     
    9. Dispositif selon la revendication 8, caractérisé en ce que l'unité d'ajustement (304) comprend :

    un premier module d'ajustement (402) configuré pour, lorsque l'on détecte qu'une charge totale de l'au moins un nœud avec l'attribut actuel de CP est supérieure à une première valeur de seuil prédéterminée et qu'une charge totale de l'au moins un nœud avec l'attribut actuel d'UP est inférieure à une deuxième valeur de seuil prédéterminée, ajuster l'attribut d'un ou plusieurs nœuds avec l'attribut actuel d'UP pour être CP ; et/ou

    un second module d'ajustement (404) configuré pour, lorsque l'on détecte que la charge totale de l'au moins un nœud avec l'attribut actuel d'UP est supérieure à une troisième valeur de seuil prédéterminée et que la charge totale de l'au moins un nœud avec l'attribut actuel de CP est inférieure à la deuxième valeur de seuil prédéterminée, ajuster l'attribut d'un ou plusieurs nœuds avec l'attribut actuel de CP pour être UP.


     
    10. Dispositif selon la revendication 9, caractérisé en ce
    qu'une valeur de rapport du nombre d'utilisateurs accédant actuellement à un système à un nombre maximal d'utilisateurs capables d'être supportés par chaque nœud avec l'attribut de CP est prise comme charge totale actuelle de l'au moins un nœud avec l'attribut de CP, dans lequel le système est un système de communication mobile ; et/ou
    que la charge totale actuelle de l'au moins un nœud avec l'attribut d'UP est déterminée conformément à une charge d'unité de traitement central CPU actuelle et d'un trafic de plan média.
     
    11. Dispositif selon la revendication 8, caractérisé en ce que les un ou plusieurs nœuds dont l'attribut est ajusté sont un ou plusieurs nœuds avec une charge actuelle minimale.
     
    12. Dispositif selon la revendication 8, caractérisé en ce que le dispositif est en outre configuré pour, avant que l'unité d'ajustement n'ajuste l'attribut des un ou plusieurs nœuds de la pluralité de nœuds, juger si un ou plusieurs nœuds restant de même attribut que les un ou plusieurs nœuds ajustés est ou sont capables de supporter, une fois l'attribut des un ou plusieurs nœuds ajusté, une charge de service correspondant à l'attribut.
     
    13. Dispositif selon l'une quelconque des revendications 8 à 12, caractérisé en ce que l'unité d'ajustement (304) comprend en outre :

    un module d'envoi (502) configuré pour envoyer une instruction d'ajustement d'attribut à chaque nœud à ajuster, dans lequel l'instruction d'ajustement d'attribut est utilisée pour instruire le nœud de changer son attribut actuel ; et

    un module de rétablissement (504) configuré pour, si un message d'indication d'achèvement d'ajustement renvoyé par le nœud n'est pas reçu dans une période prédéterminée, rétablir l'attribut du nœud à l'attribut avant ajustement.


     
    14. Dispositif selon l'une quelconque des revendications 8 à 12, caractérisé en ce que l'unité de surveillance (302) est configurée pour surveiller la pluralité de nœuds du groupe de ressources partagées de la manière suivante :

    le rapport périodique, par chaque nœud du groupe de ressources partagées, d'une charge actuelle du nœud et d'une capacité de traitement du nœud, dans lequel la capacité de traitement se réfère à la charge CP ou à la charge UP qui peut être supportée par les nœuds ; ou

    la détection périodique, par un module de gestion de configuration de ressources, de la charge actuelle et de la capacité de traitement de chaque nœud dans le groupe de ressources partagées.


     




    Drawing
























    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