WO2014190557A1 - 一种物理资源调整方法、装置及控制器 - Google Patents

一种物理资源调整方法、装置及控制器 Download PDF

Info

Publication number
WO2014190557A1
WO2014190557A1 PCT/CN2013/076602 CN2013076602W WO2014190557A1 WO 2014190557 A1 WO2014190557 A1 WO 2014190557A1 CN 2013076602 W CN2013076602 W CN 2013076602W WO 2014190557 A1 WO2014190557 A1 WO 2014190557A1
Authority
WO
WIPO (PCT)
Prior art keywords
physical resources
allocated
workload
load
board
Prior art date
Application number
PCT/CN2013/076602
Other languages
English (en)
French (fr)
Inventor
陈取才
姚建辉
Original Assignee
华为技术有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 华为技术有限公司 filed Critical 华为技术有限公司
Priority to EP13885716.4A priority Critical patent/EP2996382B1/en
Priority to PCT/CN2013/076602 priority patent/WO2014190557A1/zh
Priority to CN201380001444.0A priority patent/CN103703830B/zh
Publication of WO2014190557A1 publication Critical patent/WO2014190557A1/zh
Priority to US14/951,629 priority patent/US10142997B2/en

Links

Classifications

    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04WWIRELESS COMMUNICATION NETWORKS
    • H04W72/00Local resource management
    • H04W72/50Allocation or scheduling criteria for wireless resources
    • H04W72/54Allocation or scheduling criteria for wireless resources based on quality criteria
    • H04W72/542Allocation or scheduling criteria for wireless resources based on quality criteria using measured or perceived quality
    • GPHYSICS
    • G06COMPUTING; CALCULATING OR COUNTING
    • G06FELECTRIC DIGITAL DATA PROCESSING
    • G06F9/00Arrangements for program control, e.g. control units
    • G06F9/06Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
    • G06F9/46Multiprogramming arrangements
    • G06F9/50Allocation of resources, e.g. of the central processing unit [CPU]
    • G06F9/5083Techniques for rebalancing the load in a distributed system
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/04Network management architectures or arrangements
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/34Signalling channels for network management communication
    • H04L41/344Out-of-band transfers
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0805Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability
    • H04L43/0817Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters by checking availability by checking functioning
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/10Flow control; Congestion control
    • H04L47/12Avoiding congestion; Recovering from congestion
    • H04L47/125Avoiding congestion; Recovering from congestion by balancing the load, e.g. traffic engineering
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00Traffic control in data switching networks
    • H04L47/70Admission control; Resource allocation
    • H04L47/82Miscellaneous aspects
    • H04L47/822Collecting or measuring resource availability data
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08Configuration management of networks or network elements
    • H04L41/0803Configuration setting
    • H04L41/0813Configuration setting characterised by the conditions triggering a change of settings
    • H04L41/0816Configuration setting characterised by the conditions triggering a change of settings the condition being an adaptation, e.g. in response to network events
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/08Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
    • H04L43/0876Network utilisation, e.g. volume of load or congestion level
    • H04L43/0888Throughput
    • HELECTRICITY
    • H04ELECTRIC COMMUNICATION TECHNIQUE
    • H04LTRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L43/00Arrangements for monitoring or testing data switching networks
    • H04L43/16Threshold monitoring

Definitions

  • the present invention relates to the field of electronic technologies, and in particular, to a physical resource adjustment method, apparatus, and controller. Background technique
  • the functions of the wireless network layer can be divided into two functions: CP (Control Plane) / UP (User Plane). section.
  • the CP and the UP function are generally carried by different physical boards.
  • the board includes two types of boards: the CP board and the UP board.
  • the carrier needs to purchase a certain number of CP boards and a certain number according to the prediction.
  • UP boards it is more difficult to accurately predict CP/UP capability requirements. Therefore, when the prediction is not accurate, one type of board is wasted, and the other type of board is not enough. If the adjustment is needed, the insufficient type of board needs to be re-purchased, so that the adjustment period is made. Long; and the adjustment process requires manual intervention, which will cause business interruption. Summary of the invention
  • the embodiment of the invention provides a physical resource adjustment method and device, which can automatically adjust and allocate physical resources according to the load conditions of the CP and the UP.
  • the embodiment of the present invention provides a physical resource adjustment method, including: collecting load information of a CP from a control plane CP board, and collecting UP load information from a user plane UP board, and according to corresponding load information. Determine the CP workload and UP workload;
  • the physical resources of the CP and the UP are adjusted according to the calculated number of allocated physical resources.
  • the adjusting the physical resources of the CP and the UP according to the calculated number of the allocated physical resources includes:
  • the adjustment indication information is sent according to the calculated number of allocated physical resources, to instruct the user to adjust the physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the CP information is collected from the control plane CP board, and the user plane UP board
  • the UP load information is collected, and the CP workload and the UP workload are determined according to the corresponding load information, including:
  • the load information of the CP is collected from the CP of the control plane, and the load information of the UP is collected from the UP board of the user plane, and the corresponding CP load and the UP load are determined.
  • the load information includes the CPU usage, the number of users, and the throughput. Any one or more of the rates;
  • the maximum value of the CP load amount or the average value of the load amount in the preset time period is taken as the workload of the CP, and the maximum value of the UP load amount or the average value of the load amount in the preset time period is used as the workload of the UP.
  • the CP workload and the UP When the working load meets the preset equalization adjustment condition, the number of allocated physical resources is calculated, including: when the CP workload and the UP workload meet the preset equalization adjustment condition, determine the number of CP physical resources currently allocated for the CP and allocate for the UP. Number of UP physical resources;
  • the number of allocated physical resources is calculated based on the number of CP physical resources and the CP workload currently allocated to the CP, the number of UP physical resources currently allocated for the UP, and the UP workload.
  • the calculating the number of allocated physical resources includes:
  • Num A xALoad work Num B BLoad work
  • Num A + X Num B - X where ⁇ is the number of physical resources currently allocated for the scarce faces in CP and UP, ALoad work is the workload of the scarce faces in CP and UP, and Num B is currently The number of physical resources allocated by the redundant plane in CP and UP, ⁇ . ⁇ is the workload of redundant faces in ? and 11?, which is the number of allocated physical resources.
  • the CP workload and the UP workload satisfy the preset equalization adjustment condition:
  • the absolute value or ratio of the difference between the load and the UP workload is outside the preset range threshold.
  • the calculating according to the calculation The number of physical resources adjusts the physical resources of the CP and UP, including:
  • the logical type of the number of boards indicated by the number of allocated physical resources in the board of the CP is changed to the UP logical type to implement the UP function.
  • the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logic type to implement the CP function.
  • the calculating according to the calculation The number of physical resources adjusts the physical resources of the CP and UP, including:
  • the software configuration is changed, and the CP software instance corresponding to the number of the allocated physical resources is deleted according to the software configuration, and the deployment physical activity is newly created on the free physical resources.
  • the number of source corresponds to the number of UP software instances;
  • the software configuration is changed, and the software instance of the UP number corresponding to the number of allocated physical resources is deleted according to the software configuration, and the CP software corresponding to the number of the allocated physical resources is newly created on the idle physical resource.
  • the calculating according to the calculation The number of physical resources adjusts the physical resources of the CP and UP, including:
  • the number of physical resources used by the software instance of the CP is reduced according to the indication of the number of allocated physical resources, and the physical resources corresponding to the number of allocated physical resources are allocated to the physical resources.
  • the number of physical resources used by the UP software instance is reduced according to the indication of the number of allocated physical resources, and the physical resources corresponding to the number of idle physical resources are allocated to the CP software instance.
  • the redundant surface Refers to CP and UP, which compares the acquired CP workload with the UP workload, where the workload is smaller;
  • the scarce surface refers to the CP and UP, which compares the acquired CP workload with the UP workload, where the workload is larger.
  • Physical resources of CP and UP including:
  • the number of allocated physical resources is submitted to the operating system.
  • the OS is automatically scheduled by the OS; or, the virtualized adjustment is performed according to the number of allocated physical resources; or the cloud computing adjustment is performed according to the number of allocated physical resources.
  • the embodiment of the present invention provides a physical resource adjustment apparatus, including: a collection module, configured to collect load information of a CP from a control plane CP board, and collect UP load information from a user plane UP board, and Obtaining the CP workload and the UP workload according to the corresponding load information;
  • a calculation module configured to calculate a number of allocated physical resources when the CP workload and the UP workload meet the preset equalization adjustment condition
  • the adjustment module is configured to adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the adjusting module includes: an adjusting unit, configured to automatically adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources;
  • the indication unit is configured to send the adjustment indication information according to the calculated number of the allocated physical resources, to indicate that the user adjusts the physical resources of the CP and the UP according to the calculated number of the allocated physical resources.
  • the collecting module includes:
  • the acquiring unit is configured to collect load information of the CP from the control plane CP board, and from the user plane UP
  • the board collects load information of the UP, and determines a corresponding CP surface load amount and a UP surface load amount, where the load information includes any one or more of a CPU usage rate, a number of users, and a throughput rate;
  • the recording unit is configured to use the maximum value of the CP load amount or the average value of the load amount in the preset time period as the workload of the CP, and the maximum value of the UP load amount or the average value of the load amount in the preset time period as the UP Work load.
  • the calculating module includes: determining a unit, When the CP workload and the UP workload meet the preset equalization adjustment conditions, determine the number of CP physical resources currently allocated for the CP and the number of UP physical resources allocated for the UP; and the calculating unit, configured to allocate the CP according to the current CP The number of physical resources and the CP workload, the number of UP physical resources currently allocated for UP, and the UP workload are calculated.
  • the calculating unit specifically calculates the number of allocated physical resources, so that each adjusted CP physical resource is the total work of the current CP.
  • the absolute value or ratio of the difference between the load-sharing load value and the adjusted load value of the current UP total workload on each UP physical resource is within a preset range threshold.
  • the computing module further includes:
  • condition calculation unit configured to calculate an absolute value or a ratio of a difference between the CP workload and the UP workload
  • the determining unit is configured to determine whether the result calculated by the condition calculating unit is outside a preset range threshold, and if yes, the CP workload and the UP workload satisfy a preset equalization adjustment condition, and notify the determining unit.
  • the adjusting module is specifically used to: when the CP is a redundant interface, the board of the CP is used. The logical type of the number of boards that are allocated in the number of physical resources is changed to the UP logical type to implement the UP function. When the UP is used as the redundant interface, the number of physical resources allocated in the UP board is configured. The logical type of the indicated number of boards is modified to the CP logic type to implement the CP function.
  • the adjusting module is specifically used to change a software configuration when the CP is used as a redundant plane, according to
  • the software configuration deletes the number of the CP software instances corresponding to the number of allocated physical resources, and creates a new number of UP software instances corresponding to the number of allocated physical resources on the idle physical resources; when the UP is a redundant plane, the software configuration is changed. And deleting the software instance of the UP corresponding to the number of allocated physical resources according to the software configuration, and creating a CP software instance corresponding to the number of the allocated physical resources on the idle physical resource.
  • the adjusting module is specifically configured to: when the CP is used as a redundant plane, according to the number of allocated physical resources
  • the indication reduces the number of physical resources used by the software instance of the CP, and allocates the physical resources corresponding to the number of allocated physical resources to the UP software instance.
  • the indication is reduced according to the number of allocated physical resources.
  • the number of physical resources used by the instance of the UP software is allocated to the CP software instance for the physical resources corresponding to the number of allocated physical resources.
  • the eighth possible implementations, in the ninth possible implementation also include:
  • a determining module configured to determine a physical resource to be adjusted according to the calculated number of allocated physical resources; and a control module, configured to prohibit allocating new services to the determined physical resource to be adjusted and waiting to be deployed on the determined physical resource to be adjusted The business is terminated.
  • the embodiment of the present invention further provides a controller, including a control board, a CP board, and an UP board, where:
  • the control board is configured to collect load information of the CP from the CP board, from the UP board. Collecting the load information of the UP, and determining the CP workload and the UP workload according to the corresponding load information; calculating the number of allocated physical resources when the CP workload and the UP workload satisfy the preset equalization adjustment condition; Adjust the physical resources of the CP board and the UP board.
  • control board includes the physical resource adjustment apparatus of the second aspect.
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload.
  • FIG. 1 is a schematic flowchart of a physical resource adjustment method according to an embodiment of the present invention
  • FIG. 2 is a schematic flowchart of another physical resource adjustment method according to an embodiment of the present invention
  • FIG. 3 is a schematic flowchart of a specific physical resource adjustment method according to an embodiment of the present invention
  • FIG. 5 is a schematic diagram of adjustment in the physical resource adjustment method in FIG. 4;
  • FIG. 6 is a schematic flow chart of still another specific physical resource adjustment method according to an embodiment of the present invention.
  • FIG. 7 is a schematic diagram of adjustment in the physical resource adjustment method in FIG. 6;
  • Figure 8 is a schematic diagram of adjustment of operating system OS scheduling
  • FIG. 9 is a schematic structural diagram of a physical resource adjusting apparatus according to an embodiment of the present invention.
  • FIG. 10 is a schematic structural diagram of another physical resource adjustment apparatus according to an embodiment of the present invention.
  • FIG. 11 is a schematic structural diagram of a controller according to an embodiment of the present invention. detailed description
  • FIG. 1 is a schematic flowchart of a method for adjusting a physical resource according to an embodiment of the present invention.
  • the method according to the embodiment of the present invention may be implemented, for example, by a device such as a controller, where the controller may be, for example,
  • the BSC Base Station Controller
  • RNC Radio Network Controller
  • the method includes:
  • S101 Collecting load information of the CP from the CP board of the control plane, and collecting the load information of the UP from the user plane UP board, and calculating the CP workload and the UP workload according to the corresponding load information.
  • the CP can be used to implement the RANAP (Radio Access Network Application Part) / RNSAP (Radio Network Subsystem Application Part) / NB AP (NodeB Application Part) / Application protocol such as transport bearer signaling, responsible for bearer establishment and maintenance of RAB (Radio Access Bearer) / RB (Radio Bear) / RL (Radio Link).
  • RANAP Radio Access Network Application Part
  • RNSAP Radio Network Subsystem Application Part
  • NB AP NodeB Application Part
  • Application protocol such as transport bearer signaling, responsible for bearer establishment and maintenance of RAB (Radio Access Bearer) / RB (Radio Bear) / RL (Radio Link).
  • the CP may include a plurality of software instances and physical resources (such as a CPU, etc.) for carrying the software instances, the software instance is used to implement the control plane logic function of the radio access network; the UP may also include several software instances and The physical resource (such as a CPU, etc.) carrying the software instance is used to complete the user plane logic function of the radio access network.
  • physical resources such as a CPU, etc.
  • the load information of the control plane CP and the user plane UP may include, but is not limited to, one of information such as CPU usage, number of users, throughput, or any combination thereof.
  • the CP workload and the UP workload in the S101 can be: CP busy hour load and UP busy hour load; or CP average load and UP average load.
  • the CP busy hour load and the UP busy hour load refer to the maximum load of the CP and the UP in a period of time, for example, the maximum CPU usage, the maximum number of users, and the maximum throughput rate in the CP and UP in two days.
  • the load information may be collected and recorded in a preset time interval, and the maximum value of the load in the preset time interval may be determined, so that the CP busy time load and the UP busy time load may be obtained, for example, the pre- Set the interval to 15 minutes, and collect the load of UP and CP in the interval every 15 minutes (such as the average load of 10 CPUs of CP or UP during these 15 minutes), and the pre-collection collected within two days.
  • the maximum load in the UP and CP loads in the time interval is set as the UP busy hour load and the CP busy time load.
  • the CP average load and the UP average load refer to the average load of CP and UP over a period of time, for example, one of the average CPU usage, the average number of users, and the average throughput rate of CP and UP in two days or any combination thereof.
  • the load information may be collected and recorded in a preset time interval, and the CP and the UP load in the preset time interval are determined, and the CP and all the preset time intervals collected in the time period are adjusted.
  • the UP load is averaged, so that the average load of the CP and the average load of the UP can be obtained.
  • the preset time interval is 15 minutes
  • the load of UP and CP in the preset time interval is collected every 15 minutes, and collected in two days.
  • the UP and CP loads in all 15 minutes were averaged as the CP average load and the UP average load.
  • the CP workload and the UP workload satisfy the preset equalization adjustment condition, which may be:
  • the absolute value or ratio of the CP workload and the UP workload difference is outside the preset range threshold. That is, one of the workloads is greater than the other.
  • the two values are equal to or approximately equal to (the difference between the two values or the ratio is within the range threshold)
  • the equilibrium adjustment condition is not met, and no equalization adjustment is needed.
  • the faces with smaller workloads in CP and UP are redundant faces, and the physical resources in the redundant faces are redundant for adjustment calls to the other face, while the faces with larger workloads are scarce faces.
  • the number of allocated physical resources calculated in S102 is used to indicate the number of physical resources that are called from the redundant plane to the scarce side, such as the number of CPUs.
  • the calculating the number of allocated physical resources may be based on the workload of the redundant plane, and the unused physical resources in the redundant plane are used as the number of allocated physical resources; or the number of allocated physical resources is calculated only according to the physical resource requirements of the scarce surface. Or calculate the number of allocated physical resources according to the number of CP physical resources and CP workload allocated for the CP, the number of UP physical resources currently allocated for UP, and the UP workload.
  • S103 Adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the specific adjustment methods include, but are not limited to, automatically changing the logic type of the board, changing the software configuration, changing the binding relationship of the service to the physical resource, and submitting to the operating system OS by the OS for automatic scheduling, virtualization, and cloud computing.
  • the adjustment is done in a variety of ways.
  • the adjusting in the S103 may include: when the CP is a redundant interface, the logical type of the number of the physical boards indicated by the number of the allocated physical resources in the board of the CP is changed to the UP logical type to implement the UP.
  • Function When the UP is a redundant interface, the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logic type to implement the CP function.
  • the method may include: when the CP is a redundant interface, changing the software configuration, deleting the CP software instance corresponding to the number of the allocated physical resources according to the software configuration, and creating the corresponding number of the allocated physical resources on the idle physical resource.
  • An instance of the UP software or: When the UP is a redundant interface, the software configuration is changed, and the software instance of the number of UPs corresponding to the number of allocated physical resources is deleted according to the software configuration, and the provisioning physics is newly created on the idle physical resources.
  • the number of resources corresponds to the number of CP software instances.
  • the method may include: when the CP is a redundant plane, reducing the number of physical resources used by the software instance of the CP according to the indication of the number of allocated physical resources, and allocating the physical resources corresponding to the number of the allocated physical resources to the UP software instance.
  • the UP is used as the redundant interface, the number of physical resources used by the UP software instance is reduced according to the number of allocated physical resources, and the physical resources corresponding to the number of idle physical resources are allocated to the CP software instance.
  • the method further includes:
  • the determined allocation of new services to the physical resources to be adjusted and the termination of services that have been deployed on the determined physical resources to be adjusted are prohibited.
  • the Shutdown operation can be performed according to the calculated number of allocated physical resources, that is, the allocation of new users to the physical resources to be adjusted (which can be allocated to other physical resources that are not adjusted) is prohibited, and the old users are already deployed.
  • the users on the physical resources to be adjusted are naturally hanged up, and then adjusted and allocated.
  • the adjustment process does not interrupt the call of the existing user, and the new user can also access normally.
  • the entire adjustment process does not have any adverse impact on the terminal user.
  • the S103 in the embodiment of the present invention may specifically adjust the physical resources of the CP and the UP according to the calculated number of allocated physical resources;
  • the method further includes: sending adjustment indication information according to the calculated number of allocated physical resources, to instruct the user to adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the users are mainly operation and maintenance personnel.
  • the present invention has the following beneficial effects:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload.
  • FIG. 2 is a schematic flowchart of another physical resource adjustment method according to an embodiment of the present invention.
  • the method described in the embodiment of the present invention may be implemented by using a network controller or the like in a 3G/2G. Specifically, the method is as follows. Methods include:
  • S201 Collecting the load information of the CP from the CP board of the control plane, and collecting the load information of the UP from the user plane UP board, and calculating the CP workload and the UP workload according to the corresponding load information.
  • the load information of the CP and the UP can be actively reported by the CP and the UP card (which can be either a normalized board or a separate board).
  • the load information that is sent back to the network controller and the like by the CP and the UP board in response to the load collection request is sent by the controller and the device to the CP and the UP board.
  • the number of CP physical resources allocated for the CP and the number of UP physical resources allocated for the UP may be recorded after the network controller is completed, and the number of CP physical resources allocated by the CP and the number of UP physical resources allocated for the UP are respectively recorded. Or the number of physical resources of the recorded CP and the number of physical resources of the UP after the last physical resource adjustment.
  • S203 Calculate the number of allocated physical resources according to the number of CP physical resources and CP workload currently allocated to the CP, the number of UP physical resources currently allocated for the UP, and the UP workload.
  • calculating the number of allocated physical resources includes: calculating the allocated physical resources
  • the value of the total workload of the current CP on each of the CP physical resources adjusted according to the number of allocated physical resources is the load value of the total workload of the current UP on each UP physical resource.
  • the absolute value or ratio of the difference is within a preset range threshold.
  • the specific calculation formula for calculating the number of allocated physical resources may include:
  • Equation 1 Num A + X _ Num B -X '
  • Nwrn ⁇ is the number of physical resources allocated for the scarce faces in CP and UP
  • ALoac ⁇ ork is the workload of the scarce faces in CP and UP.
  • Num B is the number of physical resources allocated for redundant faces in CP and UP
  • S ⁇ ⁇ is?
  • the workload of the redundant side in the UP which is the number of allocated physical resources.
  • Equation 1 when CP is a scarce face, UP is a redundant face, and the workload is a busy hour load, the specific calculation formula is Equation 2 below.
  • is the number of CP physical resources currently allocated for the CP
  • CPLoad busy CP busy time load is the number of UP physical resources currently allocated for UP
  • UPLoad is UP busy time load , the number is allocated physical resources.
  • CP x CP indicates that the physical resources currently allocated by the CP are at the maximum load (busy load), the total load of the CP, and the expression on the left indicates that the CP has added X physics. After the resource, the load on the total workload for each physical resource. same.
  • Formula 1 in the expression to the right of the equal sign Xt/P oi ⁇ indicates the total load of UP when the physical resources currently allocated by UP are at the maximum load (busy load), and the expression on the right indicates that the CP has reduced X physical resources after each physical resource. The load that is shared by the total workload.
  • UP can be used as a scarce surface, and when CP is a redundant surface, the workload is a busy load.
  • the CPLoad busy of Formula 2 is the CP average load ⁇ oa vg , and UPLoad busy is replaced by the UP average load UPLoad avg .
  • calculation formula of the number of allocated physical resources in the embodiment of the present invention is not limited to the above calculation based on the maximum value of the load and the average value, and may be calculated according to other load statistics.
  • the specific adjustment methods include, but are not limited to, changing the logical type of the board, changing the software configuration, changing the binding relationship between the service and the physical resource, and submitting to the operating system OS for automatic scheduling by the OS, virtualization, and Any one or more of the cloud calculations complete the adjustment.
  • the adjustment may include: when the CP is a redundant interface, the logical type of the number of boards indicated by the number of allocated physical resources in the board of the CP is changed to the UP logical type to implement the UP function; In the case of a redundant interface, the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logic type to implement the CP function.
  • the method may include: when the CP is a redundant interface, changing the software configuration, deleting the CP software instance corresponding to the number of the allocated physical resources according to the software configuration, and creating the corresponding number of the allocated physical resources on the idle physical resource.
  • the UP software instance when the UP is a redundant interface, the software configuration is changed, and the software instance of the number of UPs corresponding to the number of physical resources is deleted according to the software configuration, and the number of the allocated physical resources is newly created on the idle physical resources.
  • a corresponding number of CP software instances when the UP is a redundant interface, changing the software configuration, deleting the CP software instance corresponding to the number of the allocated physical resources according to the software configuration, and creating the corresponding number of the allocated physical resources on the idle physical resource.
  • the method may include: when the CP is a redundant plane, reducing the number of physical resources used by the software instance of the CP according to the indication of the number of allocated physical resources, and allocating the physical resources corresponding to the number of the allocated physical resources to the UP software instance.
  • the UP is used as the redundant interface, the number of physical resources used by the UP software instance is reduced according to the number of allocated physical resources, and the physical resources corresponding to the number of idle physical resources are allocated to the CP software instance.
  • the method further includes:
  • the shutdown of the Shutdown operation is performed according to the calculated number of allocated physical resources, that is, the allocation of new users to the physical resources to be adjusted is prohibited (can be allocated to other physical resources that are not adjusted;), waiting for the old users (already deployed in these)
  • the user on the physical resource to be adjusted naturally hangs up, and then adjusts the allocation, so that the adjustment process does not interrupt the call of the existing user, and the new user can also access normally.
  • the entire adjustment process has no adverse effects on the end user.
  • the CPU usage information is used as the load information as an example.
  • Collection of CP load information and collection of UP load information For example, the CPU usage of each CPU in the time interval (for example, the average value in the time interval) can be counted in a granularity (time interval) of 15 minutes; and then all CPUs of the CP are at the time interval. The average CPU usage of the CP is taken as an average value, and the average CPU usage of the CP during the time interval is obtained, that is, the CP load CPLoad in the time interval. Similarly, the statistics UP CPU usage of each CPU in the time interval.
  • Rate for example, take the average value of the time interval
  • the UP load in the interval is UPLoad.
  • the specific calculation formula can be:
  • CP busy time load CPLoad Busy Max ⁇ CPload x , CPload 2 , ..., CP / , ..., CPload M );
  • UP busy time load UPLoad Busy MaxQJPload,, UPload 2 , ..., ⁇ , ..., UPload M );
  • CP/oi ⁇ refers to the CP load obtained from the first time interval, such as the CP load obtained in the first 15 minutes.
  • ⁇ ⁇ / ⁇ refers to the UP load obtained by the first time interval, such as the UP load obtained in the first 15 minutes.
  • the maximum value of the CP load of the M time intervals in the adjustment time period is the CP busy time load
  • the maximum value of the UP load of the M time intervals in the adjustment time period is the UP busy time load.
  • the value of M can be set as needed, that is, the M value can be set as needed to determine the adjustment timing.
  • the M value can be determined according to the time when the network controller is generally idle to determine the adjustment timing.
  • the specific calculation formula can also be:
  • CPLoad Busy k Max(CPload l , CPload 2 , , CPload 96 );
  • UPLoad Busy k Max(UPload 1 , UPload 2 , , UPload 96 );
  • CPLoad Busy Avg(CPLoad Busy l , CPLoad Busy 2 , ... , CPLoad Busy N )
  • UPLoad Busy Avg(UPLoad Busy l , UPLoad Busy 2 , ... , UPLoad Busy N );
  • UPLoad Busy J refers to the UP load on day ⁇ .
  • the value of the number of time intervals M in the adjustment time period is set by the user as needed, and the M value can be set as needed to determine the adjustment timing. For example, the user determines the M value according to the time when the network controller is generally idle to determine the adjustment timing.
  • the CP workload and the UP workload in the embodiment of the present invention are represented by the CP busy load and the UP busy load.
  • the CP workload and the UP workload can also be expressed by the average load
  • the CP average load and the UP average load are calculated as follows:
  • UP Load «v g M ⁇ ⁇ ⁇ , There are a total of M time intervals in the adjustment time period, and M CP load and UP load are obtained.
  • the CP/ofi ⁇ . is the CP load counted in the i-th time interval of the M time intervals
  • the CPLoad avg is the CP average load
  • the UPloadi is the UP load counted in the i-th time interval of the M time intervals.
  • UPU d avg is the UP average load.
  • CPLoad Busy k - UPLoad Busy k If the absolute value of CPLoad Busy k - UPLoad Busy k is small, it means that the load of CP and the load of UP are relatively close, basically balanced. If adjustment is not needed, S1 to S3 are continued. If the absolute silver ratio of CPLoad Busy k — UPLoad Busy k is greater than the preset threshold, it indicates that the load is unbalanced, while one side is relatively idle and the other side is busy. Adjustments are needed at this time.
  • the workload of the CP is large as a scarce surface, and the workload of the UP is small as a redundant surface.
  • the specific calculation formula for calculating the number of allocated physical resources may include:
  • ⁇ 2 ⁇ is the number of CP physical resources currently allocated for the CP
  • CP d busy CP busy time load refers to the average load of all physical resources, for example, 10 CPUs in the CP are busy Time Average load.
  • W3 ⁇ 4p is the number of UP physical resources currently allocated for UP
  • UPLoad is the up busy load, and the number is allocated physical resources.
  • the specific adjustment methods include, but are not limited to, changing the logical type of the board, changing the software configuration, changing the binding relationship between the service and the physical resource, and submitting to the operating system OS for automatic scheduling by the OS. And any one or more of the ways in cloud computing to complete the adjustment.
  • the present invention has the following beneficial effects:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload. And only according to the number of physical resources allocated for CP and UP and the calculated busy time load, the number of allocated physical resources can be obtained, and the bill of lading can be calculated, which is easy to implement.
  • FIG. 3 it is a schematic flowchart of a specific physical resource adjustment method according to an embodiment of the present invention.
  • the physical resource is adjusted by changing the logic type of the board.
  • the method includes:
  • S301 Collecting load information of the CP from the CP board of the control plane, and collecting the load information of the UP from the user plane UP board, and determining the CP workload and the UP workload according to the corresponding load information;
  • the board in this embodiment is a normalized board, that is, a common physical board of the CP and the UP, and the function software for implementing the logic function of the CP is simultaneously set and used to implement the UP.
  • Functional software for logic functions You can change the logical functions that a board can carry by modifying the logical type. The specific change can be to directly enable the CP logic function of the board or enable the UP logic function of the board.
  • FIG. 4 is a schematic flowchart of another specific physical resource adjustment method according to an embodiment of the present invention. In the embodiment of the present invention, adjusting physical resources is performed by changing a software configuration. Specifically, the method includes:
  • S401 Collecting load information of the CP from the CP board of the control plane, and collecting the load information of the UP from the user plane UP board, and determining the CP workload and the UP workload according to the corresponding load information;
  • the CPU resources of a part of the board are loaded into the CP software instance, and the CPU resources of other parts of the board are loaded into the UP software instance.
  • FIG. 5 it is an adjustment diagram in the physical resource adjustment method in FIG.
  • the physical resources indicated in Figure 5, for example, can be resources such as CPU, CPU core, or thread.
  • the box in Figure 5 is the UP software example and the ellipse box is the CP software instance.
  • the board has a total of 10 physical resources.
  • the initial stage is allocated to the UP and the CP.
  • the UP software instance is loaded in the CPU corresponding to the CP according to the number of allocated physical resources.
  • the CP software instance is loaded in the physical resource corresponding to the UP according to the number of the allocated physical resources.
  • the board in this embodiment may be a normalized board or a separate board.
  • the configuration software is used to define the physical resources used by the CP, and the UP is The physical resources used.
  • FIG. 6 is a schematic flowchart of another specific physical resource adjustment method according to an embodiment of the present invention.
  • adjusting physical resources is accomplished by changing a binding relationship between a service and a physical resource, and , the method includes:
  • S501 Collecting load information of the CP from the CP of the control plane, collecting the load information of the UP from the UP board of the user plane, and determining the CP workload and the UP workload according to the corresponding load information;
  • S502 Calculate the number of allocated physical resources when the CP workload and the UP workload meet the preset equalization adjustment condition;
  • FIG. 7 it is a schematic diagram of the adjustment in the physical resource adjustment method in Fig. 6.
  • the pool and CP pool used by the UP indicated in Figure 7 are physical resource pools including resources such as CPU, CPU core, or thread.
  • the box in Figure 7 is the UP software example, and the oval box is the CP software instance.
  • the CP and the UP are in the redundant plane, the physical resources in the pool used by the UP are allocated to the CP pool to carry the CP service.
  • the physical resource adjustment method may specifically be an OS automatic scheduling, as shown in FIG. 8 , which is an adjustment diagram based on operating system OS scheduling.
  • the UP and CP pool sub-shares indicated in FIG. 7 include a core such as a CPU and a CPU.
  • the physical resource pool of the resource such as a thread or a thread, is scheduled by the OS according to the number of allocated physical resources and the physical resources, and the corresponding software entity is loaded.
  • the box in Figure 7 is the UP software example, and the oval box is the CP software instance.
  • the present invention has the following advantages:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload.
  • a physical resource adjustment device of the present invention will be described in detail below.
  • FIG. 9 is a schematic structural diagram of a physical resource adjustment apparatus according to an embodiment of the present invention.
  • the physical resource adjustment apparatus in the embodiment of the present invention may be disposed in a device such as a controller, and the controller may be, for example,
  • the BSC may also be an RNC, which is not limited by the embodiment of the present invention.
  • the device can also be used as a third-party device, deployed outside the network controller to communicate with the controller.
  • the physical resource adjustment device may specifically include:
  • the collecting module 1 is configured to collect the load information of the CP from the control plane CP board, collect the UP load information from the user plane UP board, and determine the CP workload and the UP workload according to the corresponding load information.
  • the calculation module 2 is configured to calculate the number of allocated physical resources when the CP workload and the UP workload meet the preset equalization adjustment conditions.
  • the adjustment module 3 is configured to adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the CP may include a plurality of software instances and physical resources (such as a CPU, etc.) for carrying the software instances, the software instance is used to implement the control plane logic function of the radio access network; the UP may also include several software instances and The physical resource (such as a CPU, etc.) carrying the software instance is used to complete the user plane logic function of the radio access network.
  • physical resources such as a CPU, etc.
  • the load information of the control plane CP and the user plane UP may include, but is not limited to, one of information such as CPU usage, number of users, throughput, or the like, or any combination thereof.
  • the CP workload and the UP workload may refer to: CP busy hour load and UP busy hour load, for example, one of the maximum CPU usage, the maximum number of users, and the maximum throughput rate of CP and UP in two days or any of them. Combination; or CP average load and UP average load, for example, one of CPU usage average rate, average number of users, and average throughput rate in CP and UP within two days or any combination thereof.
  • the collecting module 1 can collect and record the load of the CP and the UP within a preset time interval, for example, 15 minutes, and then adjust the time period, for example, when two days arrive, and use the maximum value of the load as the CP busy time load and the UP busy time. Load, or average the CP and UP loads in all preset time intervals collected within two days of the adjustment time period to obtain the CP average load and the UP average load.
  • the collection module 1 can also collect and record load information at non-equal intervals, for example, randomly collecting the load conditions of the UP and CP at intervals of 15 minutes, 30 minutes, and the like.
  • the CP workload and the UP workload satisfy the preset equalization adjustment condition, which may be:
  • the absolute value or ratio of the CP workload and the UP workload difference is outside the preset range threshold. That is, one of the workloads is greater than the other.
  • the two values are equal to or approximately equal to (the difference between the two values or the ratio is within the range threshold)
  • the equilibrium adjustment condition is not met, and no equalization adjustment is needed.
  • the smaller workloads in the CP and UP are redundant faces, the physical resources in the redundant faces are redundant for adjustment calls to the other face, and the faces with larger workloads are scarce faces.
  • the calculation The number of allocated physical resources calculated by module 2 is specifically used to indicate the number of physical resources that are called from the redundant plane to the scarce plane, such as the number of CPUs.
  • the number of the allocated physical resources calculated by the calculation module 2 may be directly based on the workload of the redundant plane, and the unused physical resources in the redundant plane are used as the number of allocated physical resources; or only according to the physical resource requirements of the scarce surface.
  • the number of physical resources is allocated; or the number of allocated physical resources is calculated according to the number of CP physical resources and CP workload allocated for the CP, the number of UP physical resources currently allocated for UP, and the UP workload.
  • the specific adjustment modes adopted by the adjustment module 3 include, but are not limited to, automatically changing the logic type of the board, changing the software configuration, changing the binding relationship of the service to the physical resource, and submitting to the operating system OS by the OS for automatic scheduling, virtualization, and cloud.
  • the adjustment is done in either or more of the calculations.
  • the adjusting module 3 is specifically configured to: when the CP is a redundant interface, modify the logical type of the number of the physical boards indicated by the number of allocated physical resources in the board of the CP to The UP logical type is used to implement the UP function.
  • the UP is used as the redundant interface, the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logical type to implement the CP function.
  • the specific use is as follows: When the CP is used as the redundant interface, the software configuration is changed, and the CP software instance corresponding to the number of the allocated physical resources is deleted according to the software configuration, and the number of the allocated physical resources is newly created on the idle physical resources.
  • the number of UP software instances when UP is used as the redundant interface, the software configuration is changed, and the software instance of the number of UPs corresponding to the number of allocated physical resources is deleted according to the software configuration, and the allocated physical resources are newly created on the idle physical resources. Number of CP software instances.
  • the CP when the CP is used as a redundant plane, reduce the number of physical resources used by the software instance of the CP according to the indication of the number of allocated physical resources, and allocate the physical resources corresponding to the number of allocated physical resources to the UP software instance.
  • the UP is used as the redundant interface, the number of physical resources used by the UP software instance is reduced according to the number of allocated physical resources, and the physical resources corresponding to the number of idle physical resources are allocated to the CP software instance.
  • the adjustment module 3 may be an adjustment unit, configured to automatically adjust physical resources of the CP and the UP according to the calculated number of allocated physical resources.
  • the adjusting module 3 is an indicating unit, configured to send the adjustment indication information according to the calculated number of the allocated physical resources, to indicate that the user adjusts the physical resources of the CP and the UP according to the calculated number of the allocated physical resources.
  • the specific implementation of the physical resource adjustment apparatus in the embodiment of the present invention may refer to the foregoing figure.
  • FIG. 8 correspond to the description of the embodiment, and details are not described herein.
  • the present invention has the following beneficial effects:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload.
  • the physical resource adjustment apparatus in the embodiment of the present invention may be disposed in a device such as a controller, for example, It may be a BSC or an RNC, which is not limited by the embodiment of the present invention.
  • the device can also be used as a third-party device deployed outside the network controller to communicate with the controller.
  • the physical resource adjustment device may specifically include the collection module 1, the calculation module 2, and the adjustment module 3 in the foregoing embodiment of the present invention.
  • the collection module 1 in the physical resource adjustment device Specifically, it may include:
  • the acquiring unit 11 is configured to collect load information of the CP from the control plane CP board, and collect UP load information from the user plane UP board, where the load information includes any one of a CPU usage rate, a number of users, and a throughput rate. Or a variety;
  • the load information of the CP and the UP obtained by the obtaining unit 11 may be actively reported by the CP and the UP within a preset time threshold.
  • the busy hour load may be the maximum load of the CP or the UP within a preset time period.
  • the recording unit 12 is configured to use the maximum value of the CP load amount or the average value of the load amount in the preset time period as the workload of the CP, and the maximum value of the UP load amount or the average value of the load amount in the preset time period as UP workload.
  • the recording unit 12 records the busy time load of the corresponding face for calculation and use by the calculation module 2.
  • the calculating module 2 in the physical resource adjusting device may specifically include: a determining unit 21, configured to determine, when the CP workload and the UP workload meet the preset equalization adjustment condition, the CP currently allocated for the CP The number of the physical resources and the number of the UP physical resources allocated for the UP; the calculating unit 22 is configured to calculate the allocation according to the number of CP physical resources and the CP workload currently allocated for the CP, the number of UP physical resources currently allocated for the UP, and the UP workload. The number of physical resources.
  • the number of CP physical resources allocated for the CP and the number of UP physical resources allocated for the UP may be recorded after the network controller is completed, and the number of CP physical resources allocated by the CP and the number of UP physical resources allocated for the UP are respectively recorded. Or the number of physical resources of the recorded CP and the number of physical resources of the UP after the last physical resource adjustment.
  • the calculation unit 22 calculates the number of the allocated physical resources, including: the adjusted load value of the current CP for each CP physical resource and the adjusted UP physical resource.
  • the absolute value or ratio of the difference between the load values of the current UP total workload is within a preset range threshold.
  • the specific calculation formula can refer to Formula 1, Formula 2 and Formula 3 in the above method embodiment.
  • calculation module 2 may further include a lower unit for determining whether the balance adjustment condition is met, including:
  • condition calculation unit 23 configured to calculate an absolute value or a ratio of a difference between the CP workload and the UP workload
  • the determining unit 24 is configured to determine whether the result calculated by the condition calculating unit is outside a preset range threshold, and if yes, the CP workload and the UP workload satisfy a preset equalization adjustment condition, and notify the determining unit.
  • the adjusting module 3 in the physical resource adjusting device is specifically configured to: when the CP is a redundant interface, the number of the physical resources allocated in the board of the CP is indicated.
  • the logical type of the card is changed to the UP logical type to implement the UP function.
  • the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logical type. To achieve CP function. Or
  • the adjusting module 3 is specifically configured to: when the CP is used as the redundant interface, modify the logical type of the number of the physical boards indicated by the number of allocated physical resources in the board of the CP to the UP logical type to implement the UP function; As a redundant interface, the logical type of the number of boards indicated by the number of allocated physical resources in the UP board is changed to the CP logic type to implement the CP function. Or
  • the adjusting module 3 is specifically configured to: when the CP is used as a redundant plane, change the software configuration, delete the CP software instance corresponding to the number of the allocated physical resources according to the software configuration, and create the new deployment on the idle physical resource.
  • the number of physical resources corresponds to the number of UP software instances; when UP is used as a redundant surface
  • the software configuration is changed, the software instance of the number of UPs corresponding to the number of allocated physical resources is deleted according to the software configuration, and the CP software instance corresponding to the number of allocated physical resources is newly created on the idle physical resources.
  • the adjustment module 3 can have the above three physical resource adjustment functions at the same time, and can select one of the methods to adjust the physical resources according to specific needs.
  • the following modules may also be included to perform corresponding processing:
  • a determining module 4 configured to determine, according to the calculated number of allocated physical resources, a physical resource to be adjusted
  • the control module 5 is configured to prohibit the allocation of new services to the determined physical resources to be adjusted and wait for the service termination that has been deployed on the determined physical resources to be adjusted.
  • the shutdown of the Shutdown operation is performed according to the calculated number of the allocated physical resources.
  • the control module 5 prohibits the allocation of new users to the physical resources to be adjusted. Other physical resources that are not adjusted are allocated. Waiting for old users (users already deployed on these physical resources to be adjusted) to hang up naturally, and then adjust the allocation, so that the adjustment process does not interrupt the calls of existing users, new users. It can also be accessed normally, and the entire adjustment process has no adverse effects on the end user.
  • the present invention has the following advantages:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload. And only according to the number of physical resources allocated for CP and UP and the calculated busy time load, the number of allocated physical resources can be obtained, and the bill of lading can be calculated, which is easy to implement.
  • the controller of the embodiment of the present invention includes a control board 100, a CP board 200, and an UP board 300, where: the control The board 100 is configured to collect the load information of the CP from the CP board 200, collect the UP load information from the UP board 300, and determine the CP workload and the UP workload according to the corresponding load information; Load and UP workload meet preset equilibrium adjustment conditions The physical resources of the CP board 200 and the UP board 300 are adjusted according to the calculated number of allocated physical resources.
  • control board 100 of the embodiment of the present invention includes the physical resource adjustment apparatus corresponding to the embodiment of FIG. 9 and FIG. 10 described above.
  • control board 100 reference may be made to the method and the device embodiment corresponding to FIG. 1 to FIG. 9 above, and details are not described herein again.
  • the present invention has the following advantages:
  • the physical resources can be automatically allocated according to the load conditions of the CP and the UP.
  • the CP and the UP share physical resources, balance the device load, improve the device utilization, and reduce the possibility of CP and UP overload. And only according to the number of physical resources allocated for CP and UP and the calculated busy time load, the number of allocated physical resources can be obtained, and the bill of lading can be calculated, which is easy to implement.
  • the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM).

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Environmental & Geological Engineering (AREA)
  • Quality & Reliability (AREA)
  • Debugging And Monitoring (AREA)
  • Stored Programmes (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)
  • Supply And Distribution Of Alternating Current (AREA)

Abstract

本发明实施例提供了一种物理资源调整方法及装置,其中,所述方法包括:收集控制面CP和用户面UP的负荷信息,并根据对应的负荷信息计算得到CP忙时负荷和UP忙时负荷;在CP忙时负荷和UP忙时负荷满足预设的均衡调整条件时,计算调配物理资源数;根据计算得到的调配物理资源数调整CP和UP的物理资源。采用本发明,可自动地根据CP和UP的负荷情况进行物理资源的调整分配,实现了CP和UP共享物理资源,均衡设备负荷,提升了设备利用率,减少了CP和UP过载的可能性。

Description

一种物理资源调整方法、 装置及控制器 技术领域
本发明涉及电子技术领域,尤其涉及一种物理资源调整方法、装置及控制 器。 背景技术
在 UTRAN ( Universal Mobile Telecommunications System Terrestrial Radio Access Network, 陆地无线接入网)通用模型中, 无线网络层的功能可以划分 为 CP ( Control Plane, 控制面 ) /UP ( User Plane, 用户面)功能两个部分。
现有技术中, CP、 UP功能一般使用不同物理单板承载, 例如, 包括 CP 单板和 UP单板两种类型的单板, 运营商需要根据预测购买一定数量的 CP单 板和一定数量的 UP单板, 然而准确地预测 CP/UP能力需求却是比较困难的。 因此, 在预测不准时, 造成一种类型的单板存在浪费, 而另一种类型的单板却 不够用, 而且如果需要调整, 不够用的类型的单板则需要重新购买, 使得调整 的周期长; 并且调整过程需要人工干预, 会造成业务中断。 发明内容
本发明实施例提供了一种物理资源调整方法及装置, 可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配。
第一方面, 本发明实施例提供了一种物理资源调整方法, 包括: 从控制面 CP单板收集 CP的负荷信息,并从用户面 UP单板收集 UP的负 荷信息, 并根据对应的负荷信息确定得到 CP工作负荷和 UP工作负荷;
在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算调配物 理资源数;
根据计算得到的调配物理资源数调整 CP和 UP的物理资源。
结合第一方面,在第一种可能的实现方式中, 所述根据计算得到的调配物 理资源数调整 CP和 UP的物理资源, 包括:
根据计算得到的调配物理资源数自动调整 CP和 UP的物理资源; 或者包括:
根据计算得到的调配物理资源数发出调整指示信息,以指示用户按照计算 得到的调配物理资源数调整 CP和 UP的物理资源。
结合第一方面, 或者第一方面的第一种可能的实现方式中,在第二种可能 的实现方式中, 所述从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP 工作负荷, 包括:
从控制面 CP单板收集 CP的负荷信息,并从用户面 UP单板收集 UP的负 荷信息, 并确定对应的 CP负荷量和 UP负荷量, 所述负荷信息包括 CPU使用 率、 用户数、 吞吐率中的任一种或多种;
将预设时间周期内 CP负荷量中的最大值或者负荷量的平均值作为 CP的 工作负荷, 预设时间周期内 UP 负荷量中的最大值或者负荷量的平均值作为 UP的工作负荷。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式中, 在第三种可能的实现方式中, 所述在 CP工作负荷 和 UP工作负荷满足预设的均衡调整条件时, 计算调配物理资源数, 包括: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 确定当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数;
根据当前为 CP分配的 CP物理资源数和 CP工作负荷、当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
结合第一方面的第三种可能的实现方式,在第四种可能的实现方式中,所 述计算调配物理资源数包括:
计算调配物理资源数, 使得按照调配物理资源数调整后的每一个 CP物理 资源上为当前 CP的总工作负荷分摊的负荷值、 与调整后每一个 UP物理资源 上为当前 UP的总工作负荷分摊的负荷值之间的差值的绝对值或比值在一个预 设的范围阈值内。
结合第一方面的第四种可能的实现方式,在第五种可能的实现方式中,按 NumA xALoadwork = NumB BLoadwork
NumA + X NumB - X 其中, 所述 ^^^是当前为 CP 和 UP 中的稀缺面分配的物理资源数, ALoadwork是 CP和 UP中的稀缺面的工作负荷, NumB是当前为 CP和 UP中的 冗余面分配的物理资源数, ^^ 。^是 ?和11?中的冗余面的工作负荷, 所 述 是调配物理资源数。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式中, 或者第一方面的第三种可能的实现方式,或者第一 方面的第四种可能的实现方式,或者第一方面的第五种可能的实现方式,在第 六种可能的实现方式中, CP工作负荷和 UP工作负荷满足预设的均衡调整条 件是指: CP工作负荷和 UP工作负荷的差值的绝对值或比值在预设的范围阈 值之外。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式, 或者第一方面的第三种可能的实现方式, 或者第一方 面的第四种可能的实现方式, 或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式,在第七种可能的实现方式中, 所述根据计算 得到的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 将 CP的单板中所述调配物理资源数所指示数量的 单板的逻辑类型修改为 UP逻辑类型以实现 UP功能;
当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示数量的 单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式, 或者第一方面的第三种可能的实现方式, 或者第一方 面的第四种可能的实现方式, 或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式,在第八种可能的实现方式中, 所述根据计算 得到的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资 源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建所述调配物理贺 源数对应数量的 UP软件实例;
当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资 源数对应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调配物理 资源数对应数量的 CP软件实例。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式, 或者第一方面的第三种可能的实现方式, 或者第一方 面的第四种可能的实现方式, 或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式,在第九种可能的实现方式中, 所述根据计算 得到的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件实例使 用的物理资源数, 并将空闲出的调配物理资源数对应数量的物理资源划拨给
UP软件实例使用;
当 UP作为冗余面时, 按照调配物理资源数的指示减少 UP软件实例使用 的物理资源数, 并将空闲出的调配物理资源数对应数量的物理资源划拨给 CP 软件实例使用。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式,或者第一方面的第三种可能的实现方式,或者第一方 面的第四种可能的实现方式,或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式, 或者第一方面的第七种可能的实现方式, 或 者第一方面的第八种可能的实现方式, 或者第一方面的第九种可能的实现方 式, 在第十种可能的实现方式中, 所述根据计算得到的调配物理资源数调整
CP和 UP的物理资源之前, 还包括:
根据计算得到的调配物理资源数确定待调整的物理资源;
禁止向确定的待调整的物理资源上分配新业务和等待已经部署在确定的 待调整物理资源上的业务终止。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式,或者第一方面的第三种可能的实现方式,或者第一方 面的第四种可能的实现方式,或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式,在第十一种可能的实现方式中,所述冗余面 是指 CP和 UP中,将获取到的 CP工作负荷和 UP工作负荷进行比较,其中工 作负荷较小的一面;
所述稀缺面是指 CP和 UP中,将获取到的 CP工作负荷和 UP工作负荷进 行比较, 其中工作负荷较大的一面。
结合第一方面, 或者第一方面的第一种可能的实现方式, 或者第一方面的 第二种可能的实现方式, 或者第一方面的第三种可能的实现方式, 或者第一方 面的第四种可能的实现方式, 或者第一方面的第五种可能的实现方式, 或者第 一方面的第六种可能的实现方式,或者结合第一方面的第十一种可能的实现方 式,在第十二种可能的实现方式中, 所述根据计算得到的调配物理资源数调整
CP和 UP的物理资源, 包括:
将调配物理资源数提交给操作系统 OS由 OS自动调度; 或者, 根据调配 物理资源数进行虚拟化调整; 或者根据调配物理资源数进行云计算调整。
第二方面, 本发明实施例提供了一种物理资源调整装置, 包括: 收集模块, 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP 工作负荷;
计算模块, 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件 时, 计算调配物理资源数;
调整模块, 用于根据计算得到的调配物理资源数调整 CP和 UP的物理资 源。
结合第二方面, 在第一种可能的实现方式中, 所述调整模块包括: 调整单元, 用于根据计算得到的调配物理资源数自动调整 CP和 UP的物 理资源;
或者包括:
指示单元, 用于根据计算得到的调配物理资源数发出调整指示信息, 以指 示用户按照计算得到的调配物理资源数调整 CP和 UP的物理资源。
结合第二方面, 或者第一方面的第一种可能的实现方式,在第二种可能的 实现方式中, 所述收集模块包括:
获取单元, 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息, 并确定对应的 CP面负荷量和 UP面负荷量, 所述 负荷信息包括 CPU使用率、 用户数、 吞吐率中的任一种或多种;
记录单元, 用于将预设时间周期内 CP负荷量中的最大值或者负荷量的平 均值作为 CP的工作负荷, 预设时间周期内 UP负荷量中的最大值或者负荷量 的平均值作为 UP的工作负荷。
结合第二方面, 或者第二方面的第一种可能的实现方式, 或者第二方面第 二种可能的实现方式, 在第三种可能的实现方式中, 所述计算模块包括: 确定单元, 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件 时, 确定当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数; 计算单元, 用于根据当前为 CP分配的 CP物理资源数和 CP工作负荷、 当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
结合第二方面的第三种可能的实现方式,在第四种可能的实现方式中, 所 述计算单元具体计算调配物理资源数, 使得调整后的每一个 CP物理资源上为 当前 CP的总工作负荷分摊的负荷值与调整后每一个 UP物理资源上为当前 UP 的总工作负荷分摊的负荷值之间的差值的绝对值或比值在一个预设的范围阈 值内。
结合第二方面的第三种可能的实现方式,在第五种可能的实现方式中, 所 述计算模块还包括:
条件计算单元, 用于计算 CP工作负荷和 UP工作负荷差值的绝对值或比 值;
判断单元,用于判断所述条件计算单元计算得到的结果是否在预设的范围 阈值之外, 若是, 则 CP工作负荷和 UP工作负荷满足预设的均衡调整条件, 通知所述确定单元。
结合第二方面, 或者第二方面的第一种可能的实现方式, 或者第二方面的 第二种可能的实现方式, 或者第二方面第三种可能的实现方式, 或者第二方面 的第四种可能的实现方式, 或者第二方面的第五种可能的实现方式,在第六种 可能的实现方式中, 所述调整模块, 具体用于当 CP作为冗余面时, 将 CP的 单板中所述调配物理资源数所指示数量的单板的逻辑类型修改为 UP逻辑类型 以实现 UP功能; 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数 所指示数量的单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。 结合第二方面, 或者第二方面的第一种可能的实现方式, 或者第二方面的 第二种可能的实现方式, 或者第二方面第三种可能的实现方式, 或者第二方面 的第四种可能的实现方式, 或者第二方面的第五种可能的实现方式,在第七种 可能的实现方式中, 所述调整模块, 具体用于当 CP作为冗余面时, 更改软件 配置, 根据软件配置删除所述调配物理资源数对应数量的 CP软件实例, 并在 空闲出的物理资源上新建所述调配物理资源数对应数量的 UP 软件实例; 当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资源数对 应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调配物理资源数 对应数量的 CP软件实例。
结合第二方面, 或者第二方面的第一种可能的实现方式, 或者第二方面的 第二种可能的实现方式,或者第二方面第三种可能的实现方式, 或者第二方面 的第四种可能的实现方式, 或者第二方面的第五种可能的实现方式,在第八种 可能的实现方式中, 所述调整模块, 具体用于当 CP作为冗余面时, 按照调配 物理资源数的指示减少 CP的软件实例使用的物理资源数, 并将空闲出的调配 物理资源数对应数量的物理资源划拨给 UP软件实例使用; 当 UP作为冗余面 时, 按照调配物理资源数的指示减少 UP软件实例使用的物理资源数, 并将空 闲出的调配物理资源数对应数量的物理资源划拨给 CP软件实例使用。
结合第二方面, 或者第二方面的第一种可能的实现方式, 或者第二方面的 第二种可能的实现方式, 或者第二方面第三种可能的实现方式, 或者第二方面 的第四种可能的实现方式, 或第二方面的第五种可能的实现方式, 或者第二方 面的第六种可能的实现方式, 或者第二方面第七种可能的实现方式, 或者第二 方面的第八种可能的实现方式, 在第九种可能的实现方式中, 还包括:
确定模块, 用于根据计算得到的调配物理资源数确定待调整的物理资源; 控制模块,用于禁止向确定的待调整的物理资源上分配新业务和等待已经 部署在确定的待调整物理资源上的业务终止。
第三方面, 本发明实施例还提供了一种控制器, 包括控制单板、 CP单板 和 UP单板, 其中:
所述控制单板用于从所述 CP单板收集 CP的负荷信息, 从所述 UP单板 收集 UP的负荷信息, 并根据对应的负荷信息确定得到 CP工作负荷和 UP工 作负荷; 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算调 配物理资源数; 根据计算得到的调配物理资源数调整 CP单板和 UP单板的物 理资源。
结合第三方面,在第一种可能的实现方式中, 所述控制单板包括上述第二 方面的物理资源调整装置。
实施本发明实施例, 具有以下有益效果:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。 附图说明
为了更清楚地说明本发明实施例或现有技术中的技术方案,下面将对实施 例或现有技术描述中所需要使用的附图作筒单地介绍,显而易见地, 下面描述 中的附图仅仅是本发明的一些实施例,对于本领域普通技术人员来讲,在不付 出创造性劳动性的前提下, 还可以根据这些附图获得其他的附图。
图 1是本发明实施例的一种物理资源调整方法的流程示意图;
图 2是本发明实施例的另一种物理资源调整方法的流程示意图; 图 3是本发明实施例的其中一种具体的物理资源调整方法的流程示意图; 图 4是本发明实施例的其中另一种具体的物理资源调整方法的流程示意 图;
图 5是图 4中物理资源调整方法中的调整示意图;
图 6是本发明实施例的其中又一种具体的物理资源调整方法的流程示意 图;
图 7是图 6中物理资源调整方法中的调整示意图;
图 8 于操作系统 OS调度的调整示意图;
图 9是本发明实施例的一种物理资源调整装置的结构示意图;
图 10是本发明实施例的另一种物理资源调整装置的结构示意图; 图 11是本发明实施例的一种控制器的结构示意图。 具体实施方式
下面将结合本发明实施例中的附图,对本发明实施例中的技术方案进行清 楚、 完整地描述, 显然, 所描述的实施例仅仅是本发明一部分实施例, 而不是 全部的实施例。基于本发明中的实施例,本领域普通技术人员在没有作出创造 性劳动前提下所获得的所有其他实施例, 都属于本发明保护的范围。
请参见图 1 , 是本发明实施例的一种物理资源调整方法的流程示意图; 本 发明实施例所述的方法, 例如, 可以通过控制器等设备来实现, 所述控制器, 例如,可以是 BSC( Base Station Controller,基站控制器),也可以是 RNC( Radio Network Controller, 无线网络控制器), 本发明实施例并不进行限定。具体的, 所述方法包括:
S101 : 从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息,并根据对应的负荷信息计算得到 CP工作负荷和 UP工作负荷。
CP, 可以用于实现包括 RANAP ( Radio Access Network Application Part , 无线接入网络应用部分 ) /RNSAP ( Radio Network Subsystem Application Part, 无线网络子系统应用部分) /NB AP ( NodeB Application Part, 基站应用部分) / 传输承载信令等应用协议, 负责 RAB ( Radio Access Bearer, 无线接入承载 ) /RB ( Radio Bear, 无线承载 ) /RL ( Radio link, 无线链路 )等承载建立和维护。
UP, 可以用于实现终端用户应用的输出流和数据承载。
CP中可以包括若干软件实例和用于承载该若干软件实例的物理资源 (例 如 CPU等), 该软件实例用于实现无线接入网的控制面逻辑功能; UP中也可 以包括若干软件实例和用于承载该若干软件实例的物理资源(例如 CPU等), 该软件实例用于完成无线接入网的用户面逻辑功能。
控制面 CP和用户面 UP的负荷信息, 例如, 可以包括但不限于: CPU使 用率、 用户数、 吞吐率等信息之一或它们的任意组合。 而所述 S101 中的 CP 工作负荷和 UP工作负荷可以是指: CP忙时负荷和 UP忙时负荷; 或者 CP平 均负荷和 UP平均负荷。
所述 CP忙时负荷和 UP忙时负荷是指一段时间内的 CP和 UP的最大负荷, 例如, 两天内 CP和 UP中最大的 CPU使用率、 最大用户数以及最大吞吐率之 一或它们的任意组合。 在所述 S101中, 可以在预设的时间间隔内收集并记录 负荷信息, 确定在该预设时间间隔内的负荷最大值, 从而可以得到 CP忙时负 荷以及 UP忙时负荷, 例如, 该预设时间间隔可以为 15分钟, 每 15分钟收集 该时间间隔内 UP和 CP的负荷情况(如 CP或 UP的 10个 CPU在这 15分钟 内忙时的平均负荷) , 将两天内收集到的预设时间间隔内的 UP和 CP负荷中 的最大负荷作为 UP忙时负荷和 CP忙时负荷。
所述 CP平均负荷和 UP平均负荷是指一段时间内 CP和 UP的平均负荷, 例如, 两天内 CP和 UP的 CPU平均使用率、平均用户数以及平均吞吐率之一 或它们的任意组合。 在所述 S101中, 可以在预设的时间间隔内收集并记录负 荷信息, 确定在该预设时间间隔内的 CP和 UP负荷, 将调整时间周期内收集 的所有预设时间间隔内的 CP和 UP负荷取平均, 从而可以得到 CP平均负荷 以及 UP平均负荷, 例如, 以 15分钟为该预设时间间隔, 每 15分钟收集该预 设时间间隔内的 UP和 CP的负荷情况, 将两天内收集到的所有 15分钟内的 UP和 CP的负荷取平均值作为 CP平均负荷和 UP平均负荷。
S102: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算 调配物理资源数。
CP工作负荷和 UP工作负荷满足预设的均衡调整条件可以是指: CP工作 负荷和 UP工作负荷差值的绝对值或比值在预设的范围阈值之外。 即其中一个 工作负荷大于另外一个工作负荷。 其中, 两个值等于或者约等于(两值的差值 或者比值在范围阈值内) 时, 不符合均衡调整条件, 不需要作均衡调整。
CP和 UP中的工作负荷较小的面为冗余面, 冗余面中的物理资源还有冗 余可供调整调用给另一个面使用, 而工作负荷较大的面则为稀缺面。 在所述
S102 中计算的调配物理资源数用于指示从冗余面调用到稀缺面的物理资源 数, 例如 CPU的个数等。
所述计算调配物理资源数可以是直接根据冗余面的工作负荷,将冗余面中 未被使用的物理资源均作为调配物理资源数;或者仅根据稀缺面的物理资源需 求计算调配物理资源数; 或者根据为 CP分配的 CP物理资源数和 CP工作负 荷、 当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
S103: 根据计算得到的调配物理资源数调整 CP和 UP的物理资源。 具体的调整方式包括但不限于自动更改单板逻辑类型, 更改软件配置, 更 改业务到物理资源的绑定关系, 提交给操作系统 OS由 OS自动调度, 虚拟化 以及云计算中的任一种或多种方式完成调整。
其中具体的, 所述 S103中的调整可以包括: 当 CP作为冗余面时, 将 CP 的单板中所述调配物理资源数所指示数量的单板的逻辑类型修改为 UP逻辑类 型以实现 UP功能; 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源 数所指示数量的单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。
或者可以包括: 当 CP作为冗余面时, 更改软件配置, 根据软件配置删除 所述调配物理资源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建 所述调配物理资源数对应数量的 UP软件实例; 或者: 当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资源数对应数量的 UP的软件 实例, 并在空闲出的物理资源上新建所述调配物理资源数对应数量的 CP软件 实例。
或者可以包括: 当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件实例使用的物理资源数, 并将空闲出的调配物理资源数对应数量的 物理资源划拨给 UP软件实例使用; 当 UP作为冗余面时, 按照调配物理资源 数的指示减少 UP软件实例使用的物理资源数, 并将空闲出的调配物理资源数 对应数量的物理资源划拨给 CP软件实例使用。
进一步的, 在进行上述 S103的调整执行前, 还包括:
根据计算得到的调配物理资源数确定待调整的物理资源;
禁止所述确定的往待调整的物理资源上分配新业务和等待已经部署在所 述确定的待调整物理资源上的业务终止。
即: 还可以根据计算得到的调配物理资源数执行关闭 Shutdown操作, 即 禁止往待调整的物理资源上分配新用户(可以往其他不会调整的物理资源上分 配), 等待老用户 (已经部署在这些待调整物理资源上的用户) 自然挂机, 然 后再进行调整分配, 这样调整过程不会中断已有用户的通话,新用户还可正常 接入, 整个调整过程对终端用户没有任何不良的影响。
另外, 本发明实施例中的所述 S103具体可以为根据计算得到的调配物理 资源数自动调整 CP和 UP的物理资源; 当然在其他实施例中,也可以包括: 根据计算得到的调配物理资源数发出 调整指示信息, 以指示用户按照计算得到的调配物理资源数调整 CP和 UP的 物理资源。 其中的用户主要为运营维护人员。
通过上述实施例的描述可知, 本发明具有以下有益效果:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。
再请参见图 2, 是本发明实施例的另一种物理资源调整方法的流程示意 图; 发明实施例所述的方法可以通过 3G/2G 中的网络控制器等设备来实现, 具体的, 所述方法包括:
S201: 从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息,并根据对应的负荷信息计算得到 CP工作负荷和 UP工作负荷。
本法实施例中关于所述 CP、 UP以及 CP工作负荷和 UP工作负荷的描述 参见上述实施例中 S101中的描述, 在此不赘述。
CP和 UP的负荷信息可以是 CP和 UP单板 (可以是归一化单板, 也可以 是分开的 2种单板)主动上报的。 也可以是控制器等设备向 CP和 UP单板发 起负荷收集请求后, 由所述 CP和 UP单板响应该负荷收集请求向网络控制器 等设备反馈的负荷信息。
S202: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 确定 当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数。
先判断 CP工作负荷和 UP工作负荷是否均衡, 如果不均衡, 即如果一个 面的工作负荷较大, 一个面的工作负荷较小, 则可以认为满足均衡调整条件。
所述的为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数可以在 构建网络控制器完成后, 记录的分别为 CP分配的 CP物理资源数和为 UP分 配的 UP物理资源数, 或者是在上一次物理资源调整后, 记录的 CP的物理资 源数和 UP的物理资源数。
S203: 才 据当前为 CP分配的 CP物理资源数和 CP工作负荷、 当前为 UP 分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
在本发明实施例中,计算调配物理资源数的计算包括: 计算调配物理资源 数, 使得按照调配物理资源数调整后的每一个 CP物理资源上为当前 CP的总 工作负荷分摊的负荷值与调整后每一个 UP物理资源上为当前 UP的总工作负 荷分摊的负荷值之间的差值的绝对值或比值在一个预设的范围阈值内。
具体的, 所述计算调配物理资源数具体计算公式可以包括:
NumA xALoadwork = NumB xBLoadwork
么式 1 : NumA + X _ NumB -X ' 其中,所述 Nwrn^是为 CP和 UP中的稀缺面分配的物理资源数, ALoac^ork 是 CP和 UP中的稀缺面的工作负荷, NumB是为 CP和 UP中的冗余面分配的 物理资源数, S ^ ^是 ?和 UP中的冗余面的工作负荷, 所述 是调配物 理资源数。
在公式 1 中, 当 CP为稀缺面, UP为冗余面, 且工作负荷是指忙时负荷 时, 具体计算公式为下述的公式 2。
J^ 2 > NumCP x CPLoadbusy = NumUP x UPLoadbusy
' NumCP + X NumUP― X , 其中, 所述 ^是当前为 CP分配的 CP物理资源数, CPLoadbusy CP忙时负荷, 是当前为 UP分配的 UP物理资源数, UPLoad 是 UP 忙时负荷, 所述 是调配物理资源数。
上述公式 1中, 等号左边的表达式中 CP x CP 表示 CP当前分 配的物理资源都处于最大负荷(忙时负荷) 时, CP的总负荷, 左边的表达式 表示了 CP增加了 X个物理资源后,每一个物理资源上对总工作负荷所分摊的 负荷。 同样。 公式 1等号右边的表达式中
Figure imgf000014_0001
xt/P oi^^表示了 UP当前 分配的物理资源都处于最大负荷(忙时负荷) 时, UP的总负荷, 右边的表达 式表示了 CP减少了 X个物理资源后,每一个物理资源上对总工作负荷所分摊 的负荷。 同理可以得到 UP为稀缺面, CP为冗余面时, 工作负荷为忙时负荷 即将公式 2 的 CPLoadbusy替狹为 CP 平均负荷 ^oa vg , 而将 UPLoadbusy替换为 UP平均负荷 UPLoadavg
需要说明的是,本发明实施例的调配物理资源数的计算公式不限于上述的 根据负荷最大值、 平均值计算得到, 还可以根据其他的负荷统计值计算。
得到上述的调配物理资源数 后,具体的调整方式包括但不限于更改单板 逻辑类型, 更改软件配置, 更改业务到物理资源的绑定关系, 提交给操作系统 OS由 OS自动调度, 虚拟化以及云计算中的任一种或多种方式完成调整。
其中具体的, 调整可以包括: 当 CP作为冗余面时, 将 CP的单板中所述 调配物理资源数所指示数量的单板的逻辑类型修改为 UP逻辑类型以实现 UP 功能; 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示数量 的单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。
或者可以包括: 当 CP作为冗余面时, 更改软件配置, 根据软件配置删除 所述调配物理资源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建 所述调配物理资源数对应数量的 UP软件实例; 当 UP作为冗余面时, 更改软 件配置, 根据软件配置删除所述调配物理资源数对应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调配物理资源数对应数量的 CP软件实例。
或者可以包括: 当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件实例使用的物理资源数, 并将空闲出的调配物理资源数对应数量的 物理资源划拨给 UP软件实例使用; 当 UP作为冗余面时, 按照调配物理资源 数的指示减少 UP软件实例使用的物理资源数, 并将空闲出的调配物理资源数 对应数量的物理资源划拨给 CP软件实例使用。
进一步的, 在进行上述 S203的调整执行前, 还包括:
根据计算得到的调配物理资源数确定待调整的物理资源;
禁止向确定的待调整的物理资源上分配新业务和等待已经部署在确定的 待调整物理资源上的业务终止。
即: 根据计算得到的调配物理资源数执行关闭 Shutdown操作, 即禁止往 待调整的物理资源上分配新用户 (可以往其他不会调整的物理资源上分配;), 等待老用户 (已经部署在这些待调整物理资源上的用户) 自然挂机, 然后再进 行调整分配, 这样调整过程不会中断已有用户的通话, 新用户还可正常接入, 整个调整过程对终端用户没有任何不良的影响。
具体的, 以 CPU使用率信息作为负荷信息为例进行说明。
51 , CP负荷信息的收集和 UP负荷信息的收集。 例如, 具体可以以 15分 钟为粒度(时间间隔),统计 CP每个 CPU在该时间间隔内的 CPU占用率(例 如, 取该时间间隔内的平均值 ) ; 再将 CP全部 CPU在该时间间隔内的 CPU 占用率取一个平均值, 得到 CP在该时间间隔内的平均 CPU 占用率, 即在该 时间间隔内的 CP负荷 CPLoad , 类似的, 统计 UP每个 CPU在该时间间隔内 的 CPU占用率(例如, 取该时间间隔内的平均值) ; 再将 UP全部 CPU在该 时间间隔内的 CPU占用率取一个平均值,得到 UP在该时间间隔内的平均 CPU 占用率, 即在该时间间隔内的 UP负荷 UPLoad。
52, 计算调整时间周期内 CP忙时负荷和 UP忙时负荷。
具体计算公式可以为:
CP忙时负荷 CPLoadBusy = Max{CPloadx , CPload2 ,……, CP/ ,……, CPloadM ); UP忙时负荷 UPLoadBusy = MaxQJPload,, UPload2,…… ,ΙΙΡΙοα^,……, UPloadM ); CP/oi^是指第 次时间间隔统计得到的 CP 负荷, 例如上述的第 个 15 分钟得到的 CP负荷。 ί Ρ/οί^是指第 次时间间隔统计得到的 UP负荷, 例如 上述的第 个 15分钟得到的 UP负荷。在调整时间周期内 M个时间间隔的 CP 负荷中的最大值即为 CP忙时负荷,在调整时间周期内 M个时间间隔的 UP负 荷中的最大值即为 UP忙时负荷。 M的值可以根据需要设定, 即可以根据需要 设置 M值以确定调整时机, 例如可以根据网络控制器一般处于空闲时的时间 确定 M值以确定调整时机。
或者, 具体的计算公式还可以为:
CP第 K天的忙时负荷:
CPLoad Busy k = Max(CPloadl , CPload2 , , CPload96 );
UP第 K天的忙时负荷:
UPLoad Busy k = Max(UPload1 , UPload2 , , UPload96 );
然后计算调整周期内总的忙时负荷作为 CP忙时负荷和 UP忙时负荷: CPLoadBusy = Avg(CPLoadBusy l , CPLoad Busy 2 , ... , CPLoad Busy N ) UPLoadBusy = Avg(UPLoadBusy l , UPLoad Busy 2 , ... , UPLoad Busy N); 是指第 κ天的 CP负荷, UPLoadBusy J是指第 κ天的 UP 负荷。 调整时间周期内的时间间隔数 M的值由用户根据需要设定, 可以根据 需要设置 M值以确定调整时机, 例如用户根据网络控制器一般处于空闲时的 时间确定 M值以确定调整时机。
需要说明的是, 本发明实施例的 CP工作负荷和 UP工作负荷用 CP忙时 负荷和 UP忙时负荷表示,在其他实施例中, CP工作负荷和 UP工作负荷也可 以以平均负荷表示时, CP平均负荷和 UP平均负荷的计算如下:
_ CPloadl + CPload2 + + CPloadi + + CPloadM
LPLoadavg = ― ; τ τητ , UPload, + UPload + ...... + UPload, + ...... + UPloadM
UP Load «vg = M ■ ― ·, 在调整时间周期内一共有 M个时间间隔, 获取到 M个 CP负荷以及 UP 负荷。 所述的 CP/ofi^.为 M个时间间隔中第 i个时间间隔统计到的 CP负荷, CPLoadavg为 CP平均负荷, UPloadi为 M个时间间隔中第 i个时间间隔统 计到的 UP负荷, UPU davg为 UP平均负荷。
53 , 判断负荷是否均衡。
如果 CPLoadBusy k - UPLoadBusy k的绝对值较小, 说明 CP的负荷和 UP的 负荷较为接近, 基本是均衡的, 不需要进行调整, 则继续执行 S1至 S3。 如象 CPLoad Busy k— UPLoad Busy k的绝对银较 , 大于预设的阈值,则表明 负荷不均衡, 一边比较空闲, 另一边则比较繁忙。 此时需要进行调整。
54, 计算调整量即调配物理资源数。
本发明实施例中, 以 CP的工作负荷较大作为稀缺面, UP的工作负荷较 小作为冗余面进行说明。 所述计算调配物理资源数具体计算公式可以包括:
NumCP X CPLoadbusy _ NumUP x UPLoadbusy
Numr,D + X Num 其中, 所述 ^^2^是当前为 CP分配的 CP物理资源数, CP dbusy CP忙时负荷, 是指所有物理资源的平均负荷, 例如 CP中 10个 CPU忙时的 平均负荷。 w¾p是当前为 UP分配的 UP物理资源数, UPLoad 是 up忙 时负荷, 所述 是调配物理资源数。
S5 , 调整物理资源。
计算得到上述的调配物理资源数 后,具体的调整方式包括但不限于更改 单板逻辑类型, 更改软件配置, 更改业务到物理资源的绑定关系, 提交给操作 系统 OS由 OS自动调度,虚拟化以及云计算中的任一种或多种方式完成调整。
通过上述实施例的描述可知, 本发明具有以下有益效果:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。 并且仅需根据为 CP和 UP分配的物理资源数和计算得到的忙 时负荷即可得到调配物理资源数, 计算筒单, 易于实现。
再请参见图 3 , 是本发明实施例的其中一种具体的物理资源调整方法的流 程示意图; 在本发明实施例中,调整物理资源是通过更改单板的逻辑类型来完 成调整, 具体的, 所述方法包括:
S301: 从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP工作负荷;
S302: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算 调配物理资源数;
S303 : 当 CP作为冗余面时, 将 CP的单板中所述调配物理资源数所指示 数量的单板的逻辑类型修改为 UP逻辑类型, 加载 UP软件, 用以实现 UP功 h ·
S304: 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示 数量的单板的逻辑类型修改为 CP逻辑类型, 加载 CP软件, 用以实现 CP功 h
匕。
可以理解的是, 本实施例中的所述单板为归一化单板, 即 CP和 UP的共 物理单板, 其上同时设置了用于实现 CP逻辑功能的功能软件和用于实现 UP 逻辑功能的功能软件。通过修改逻辑类型的方式可以更改单板可以承载的逻辑 功能。 具体的更改可以为直接启用单板的 CP逻辑功能或者启用单板的 UP逻 辑功能。 图 4, 是本发明实施例的其中另一种具体的物理资源调整方法的流程示意 图; 在本发明实施例中, 调整物理资源是通过更改软件配置来完成调整, 具体 的, 所述方法包括:
S401: 从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP工作负荷;
S402: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算 调配物理资源数;
S403: 当 CP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配 物理资源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建所述调配 物理资源数对应数量的 UP软件实例;
S404: 当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配 物理资源数对应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调 配物理资源数对应数量的 CP软件实例。
比如, 根据调配物理资源数, 将单板中一部分的 CPU资源加载 CP软件 实例, 而将单板中其它部分的 CPU资源加载 UP软件实例。 具体如图 5所示, 是图 4中物理资源调整方法中的调整示意图。 图 5中所指出物理资源, 例如, 可以为 CPU, CPU的核、 或线程等资源。 图 5中方框为 UP软件实例, 椭圆 框为 CP软件实例。 如图 5所示, 单板一共有 10个物理资源, 初始阶段平均 分配给 UP和 CP, 当 CP忙、 UP作为冗余面时, 根据调配物理资源数在 CP 对应的 CPU中加载 UP软件实例。 当 UP忙而 CP作为冗余面时, 根据调配物 理资源数在 UP对应的物理资源中加载 CP软件实例。
可以理解的是, 本实施例中的所述单板可以为归一化单板,也可以为分开 的两种单板, 本实施例通过配置软件来定义给 CP使用的物理资源, 以及给 UP使用的物理资源。
图 6, 是本发明实施例的其中又一种具体的物理资源调整方法的流程示意 图; 在本发明实施例中,调整物理资源是通过更改业务到物理资源的绑定关系 来完成调整, 具体的, 所述方法包括:
S501 : 从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP工作负荷; S502: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算 调配物理资源数;
S503: 当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件 实例使用的物理资源数,并将空闲出的调配物理资源数对应数量的物理资源划 拨给 UP软件实例使用;
S504: 当 UP作为冗余面时, 按照调配物理资源数的指示减少 UP软件实 例使用的物理资源数,并将空闲出的调配物理资源数对应数量的物理资源划拨 给 CP软件实例使用。
比如, 对于 x%的计算资源用于调度 CP软件实例, ( 100-x ) %的计算资 源用于调度 UP软件进程。 具体图 7所示, 是图 6中物理资源调整方法中的调 整示意图。 图 7中所指出的 UP使用的池子和 CP池子是指包括如 CPU、 CPU 的核、 或线程等资源的物理资源池。 图 7 中方框为 UP软件实例, 椭圆框为 CP软件实例。 在 CP, UP作为冗余面时, 将 UP使用的池子中的物理资源调 度到 CP池子中, 承载 CP业务。
进一步的, 物理资源调整方法具体还可以为 OS自动调度, 具体如图 8所 示, 是基于操作系统 OS调度的调整示意图, 图 7中所指出的 UP和 CP池子 共用包括如 CPU、 CPU的核、 或线程等资源的物理资源池, 由 OS根据调配 物理资源数, 以及物理资源的能力进行调度, 加载相应面的软件实体。 图 7 中方框为 UP软件实例, 椭圆框为 CP软件实例。
通过上述实施例的描述可知, 本发明具有以下优点:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。
下面对本发明的一种物理资源调整装置进行详细说明。
请参见图 9, 是本发明实施例的一种物理资源调整装置的结构示意图, 本 发明实施例的所述物理资源调整装置可以设置在控制器等设备中, 所述控制 器, 例如, 可以是 BSC, 也可以是 RNC, 本发明实施例并不进行限定。 所述 装置也可以作为第三方设备,部署在网络控制器之外与控制器通信连接。所述 的物理资源调整装置具体可以包括: 收集模块 1 , 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP 工作负荷。
计算模块 2, 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条 件时, 计算调配物理资源数。
调整模块 3 , 用于根据计算得到的调配物理资源数调整 CP和 UP的物理 资源。
CP中可以包括若干软件实例和用于承载该若干软件实例的物理资源 (例 如 CPU等), 该软件实例用于实现无线接入网的控制面逻辑功能; UP中也可 以包括若干软件实例和用于承载该若干软件实例的物理资源(例如 CPU等), 该软件实例用于完成无线接入网的用户面逻辑功能。
控制面 CP和用户面 UP的负荷信息, 例如, 可以包括但不限于: CPU使 用率、 用户数、吞吐率等信息之一或它们的任意组合。 而所述 CP工作负荷和 UP工作负荷可以是指: CP忙时负荷和 UP忙时负荷, 例如两天内 CP和 UP 中最大的 CPU使用率、 最大用户数以及最大吞吐率之一或它们的任意组合; 或者 CP平均负荷和 UP平均负荷, 例如, 两天内 CP和 UP中 CPU使用平均 率、 平均用户数以及平均吞吐率之一或它们的任意组合。
所述收集模块 1可以在预设的时间间隔例如 15分钟内收集并记录 CP和 UP的负荷, 然后调整时间周期例如两天到来时,将其中的负荷最大值作为 CP 忙时负荷以及 UP忙时负荷, 或者将调整时间周期两天内收集的所有预设时间 间隔内的 CP和 UP负荷取平均,从而得到 CP平均负荷和 UP平均负荷。所述 收集模块 1也可以非等时间间隔内收集并记录负荷信息, 例如, 随机地以 15 分钟、 30分钟等时间间隔收集 UP和 CP的负荷情况。
CP工作负荷和 UP工作负荷满足预设的均衡调整条件可以是指: CP工作 负荷和 UP工作负荷差值的绝对值或比值在预设的范围阈值之外。 即其中一个 工作负荷大于另外一个工作负荷。 其中, 两个值等于或者约等于(两值的差值 或比值在范围阈值内) 时, 不符合均衡调整条件, 不需要作均衡调整。
CP和 UP中的工作负荷较小的面为冗余面, 冗余面中的物理资源还有冗 余可供调整调用给另一个面使用, 而工作负荷较大的面则为稀缺面。所述计算 模块 2 计算的调配物理资源数具体用于指示从冗余面调用到稀缺面的物理资 源数, 例如 CPU的个数等。
所述计算模块 2 计算的调配物理资源数可以是直接根据冗余面的工作负 荷,将冗余面中未被使用的物理资源均作为调配物理资源数; 或者仅根据稀缺 面的物理资源需求计算调配物理资源数; 或者根据为 CP分配的 CP物理资源 数和 CP工作负荷、 当前为 UP分配的 UP物理资源数和 UP工作负荷,计算调 配物理资源数。
所述调整模块 3 所采用的具体调整方式包括但不限于自动更改单板逻辑 类型, 更改软件配置, 更改业务到物理资源的绑定关系, 提交给操作系统 OS 由 OS自动调度, 虚拟化以及云计算中的任一种或多种方式完成调整。
其中具体的, 所述调整模块 3在调整物理资源时, 具体用于: 当 CP作为 冗余面时, 将 CP的单板中所述调配物理资源数所指示数量的单板的逻辑类型 修改为 UP逻辑类型以实现 UP功能; 当 UP作为冗余面时,将 UP的单板中所 述调配物理资源数所指示数量的单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。
或者具体用于: 当 CP作为冗余面时, 更改软件配置, 根据软件配置删除 所述调配物理资源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建 所述调配物理资源数对应数量的 UP软件实例; 当 UP作为冗余面时, 更改软 件配置, 根据软件配置删除所述调配物理资源数对应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调配物理资源数对应数量的 CP软件实例。
或者具体用于: 当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件实例使用的物理资源数, 并将空闲出的调配物理资源数对应数量的 物理资源划拨给 UP软件实例使用; 当 UP作为冗余面时, 按照调配物理资源 数的指示减少 UP软件实例使用的物理资源数, 并将空闲出的调配物理资源数 对应数量的物理资源划拨给 CP软件实例使用。
在本实施例中, 所述调整模块 3具体可以为一调整单元, 用于根据计算得 到的调配物理资源数自动调整 CP和 UP的物理资源。 或者, 所述调整模块 3 为一指示单元, 用于根据计算得到的调配物理资源数发出调整指示信息, 以指 示用户按照计算得到的调配物理资源数调整 CP和 UP的物理资源。 具体的,本发明实施例的所述物理资源调整装置的具体实现可参照上述图
1至图 8对应实施例的描述, 在此不赘述。
通过上述实施例的描述可知, 本发明具有以下有益效果:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。
再请参见图 10, 是本发明实施例的另一种物理资源调整装置的结构示意 图, 本发明实施例的所述物理资源调整装置可以设置在控制器等设备中, 所述 控制器, 例如, 可以是 BSC, 也可以是 RNC, 本发明实施例并不进行限定。 所述装置也可以作为第三方设备, 部署在网络控制器之外与控制器通信连接。 所述的物理资源调整装置具体可以包括上述图 9对应实施例中的收集模块 1、 计算模块 2以及调整模块 3 , 在本发明实施例中, 所述物理资源调整装置中的 所述收集模块 1具体可以包括:
获取单元 11 , 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息, 所述负荷信息包括 CPU使用率、 用户数、 吞吐 率中的任一种或多种;
所述获取单元 11获取的 CP和 UP的负荷信息可以是 CP和 UP在预设时 间阈值内主动上报的。 而所述忙时负荷则可以为在预设的时间周期内 CP或者 UP的最大负荷。
记录单元 12,用于将预设时间周期内 CP负荷量中的最大值或者负荷量的 平均值作为 CP的工作负荷, 预设时间周期内 UP负荷量中的最大值或者负荷 量的平均值作为 UP的工作负荷。
所述记录单元 12记录得到相应面的忙时负荷供所述计算模块 2计算使用。 进一步的, 所述物理资源调整装置中的所述计算模块 2具体可以包括: 确定单元 21 , 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条 件时, 确定当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数; 计算单元 22,用于根据当前为 CP分配的 CP物理资源数和 CP工作负荷、 当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
在 CP忙时负荷和 UP忙时负荷不均衡时, 即一个面的忙时负荷较大, 一 个面的忙时负荷较小, 则可以认为满足均衡调整条件, 需要进行调整。
所述的为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数可以在 构建网络控制器完成后, 记录的分别为 CP分配的 CP物理资源数和为 UP分 配的 UP物理资源数, 或者是在上一次物理资源调整后, 记录的 CP的物理资 源数和 UP的物理资源数。
在本实施例中, 所述计算单元 22计算所述调配物理资源数的原则包括: 调整后每一个 CP物理资源上为当前 CP的总工作负荷分摊的负荷值与调整后 每一个 UP物理资源上为当前 UP的总工作负荷分摊的负荷值之间的差值的绝 对值或比值在一个预设的范围阈值内。其具体的计算公式可以参照上述方法实 施例中的公式 1、 公式 2以及公式 3。
进一步的,所述计算模块 2还可以包括一下单元用于确定是否满足均衡调 整条件, 包括:
条件计算单元 23 , 用于计算 CP工作负荷和 UP工作负荷差值的绝对值或 比值;
判断单元 24, 用于判断所述条件计算单元计算得到的结果是否在预设的 范围阈值之外, 若是, 则 CP工作负荷和 UP工作负荷满足预设的均衡调整条 件, 通知所述确定单元。
进一步的,在本发明实施例中, 所述物理资源调整装置中的所述调整模块 3 , 具体用于当 CP作为冗余面时, 将 CP的单板中所述调配物理资源数所指示 数量的单板的逻辑类型修改为 UP逻辑类型以实现 UP功能; 当 UP作为冗余 面时, 将 UP的单板中所述调配物理资源数所指示数量的单板的逻辑类型修改 为 CP逻辑类型以实现 CP功能。 或者
所述调整模块 3 , 具体用于当 CP作为冗余面时, 将 CP的单板中所述调 配物理资源数所指示数量的单板的逻辑类型修改为 UP逻辑类型以实现 UP功 能; 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示数量的 单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。 或者
所述调整模块 3 , 具体用于当 CP作为冗余面时, 更改软件配置, 根据软 件配置删除所述调配物理资源数对应数量的 CP软件实例, 并在空闲出的物理 资源上新建所述调配物理资源数对应数量的 UP软件实例; 当 UP作为冗余面 时, 更改软件配置, 根据软件配置删除所述调配物理资源数对应数量的 UP的 软件实例, 并在空闲出的物理资源上新建所述调配物理资源数对应数量的 CP 软件实例。
当然, 所述调整模块 3具体可以同时具有上述三种物理资源调整功能, 能 够根据具体需要选择其中一种方式进行物理资源的调整。
进一步的, 在本发明实施例中, 在所述调整模块 3进行调整之前, 还可以 包括以下模块进行相应的处理:
确定模块 4, 用于根据计算得到的调配物理资源数确定待调整的物理资 源;
控制模块 5, 用于禁止向确定的待调整的物理资源上分配新业务和等待已 经部署在确定的待调整物理资源上的业务终止。
即: 根据计算得到的调配物理资源数执行关闭 Shutdown操作, 通过所述 确定模块 4确定出待调整的物理资源后,由所述控制模块 5禁止往待调整的物 理资源上分配新用户 (可以往其他不会调整的物理资源上分配), 等待老用户 (已经部署在这些待调整物理资源上的用户 )自然挂机,然后再进行调整分配, 这样调整过程不会中断已有用户的通话,新用户还可正常接入, 整个调整过程 对终端用户没有任何不良的影响。
具体的,本发明实施例的所述物理资源调整装置的具体实现可参照上述图 1至图 8对应实施例的描述, 在此不赘述。
通过上述实施例的描述可知, 本发明具有以下优点:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。 并且仅需根据为 CP和 UP分配的物理资源数和计算得到的忙 时负荷即可得到调配物理资源数, 计算筒单, 易于实现。
再请参见图 11 , 是本发明实施例的一种控制器的结构示意图, 本发明实 施例的所述控制器包括控制单板 100、 CP单板 200和 UP单板 300 , 其中: 所述控制单板 100用于从所述 CP单板 200收集 CP的负荷信息, 从所述 UP单板 300收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负 荷和 UP工作负荷; 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件 时, 计算调配物理资源数; 根据计算得到的调配物理资源数调整所述 CP单板 200和 UP单板 300的物理资源。
具体的, 本发明实施例的所述控制单板 100包括上述图 9和图 10对应实 施例的物理资源调整装置。所述控制单板 100的具体实现可参照上述图 1至图 9对应的方法以及装置实施例, 再次不赘述。
通过上述实施例的描述可知, 本发明具有以下优点:
可自动地根据 CP和 UP的负荷情况进行物理资源的调整分配,实现了 CP 和 UP共享物理资源, 均衡设备负荷, 提升了设备利用率, 减少了 CP和 UP 过载的可能性。 并且仅需根据为 CP和 UP分配的物理资源数和计算得到的忙 时负荷即可得到调配物理资源数, 计算筒单, 易于实现。
本领域普通技术人员可以理解实现上述实施例方法中的全部或部分流程, 是可以通过计算机程序来指令相关的硬件来完成,所述的程序可存储于一计算 机可读取存储介质中,该程序在执行时,可包括如上述各方法的实施例的流程。 其中,所述的存储介质可为磁碟、光盘、只读存储记忆体(Read-Only Memory, ROM )或随机存储记忆体(Random Access Memory, RAM )等。
以上所揭露的仅为本发明一种较佳实施例而已, 当然不能以此来限定本 发明之权利范围, 因此依本发明权利要求所作的等同变化,仍属本发明所涵盖 的范围。

Claims

权 利 要 求
1、 一种物理资源调整方法, 其特征在于, 包括:
从控制面 CP单板收集 CP的负荷信息,并从用户面 UP单板收集 UP的负 荷信息, 并根据对应的负荷信息确定得到 CP工作负荷和 UP工作负荷;
在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算调配物 理资源数;
根据计算得到的调配物理资源数调整 CP和 UP的物理资源。
2、 如权利要求 1所述的方法, 其特征在于, 所述根据计算得到的调配物 理资源数调整 CP和 UP的物理资源, 包括:
根据计算得到的调配物理资源数自动调整 CP和 UP的物理资源; 或者包括:
根据计算得到的调配物理资源数发出调整指示信息,以指示用户按照计算 得到的调配物理资源数调整 CP和 UP的物理资源。
3、 如权利要求 1或 2所述的方法, 其特征在于, 所述从控制面 CP单板 收集 CP的负荷信息, 并从用户面 UP单板收集 UP的负荷信息, 并根据对应 的负荷信息确定得到 CP工作负荷和 UP工作负荷, 包括:
从控制面 CP单板收集 CP的负荷信息,并从用户面 UP单板收集 UP的负 荷信息, 并确定对应的 CP负荷量和 UP负荷量, 所述负荷信息包括 CPU使用 率、 用户数、 吞吐率中的任一种或多种;
将预设时间周期内 CP负荷量中的最大值或者负荷量的平均值作为 CP的 工作负荷, 预设时间周期内 UP 负荷量中的最大值或者负荷量的平均值作为 UP的工作负荷。
4、 如权利要求 1至 3任一项所述的方法, 其特征在于, 所述在 CP工作 负荷和 UP工作负荷满足预设的均衡调整条件时,计算调配物理资源数,包括: 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 确定当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数;
根据当前为 CP分配的 CP物理资源数和 CP工作负荷、当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
5、 如权利要求 4所述的方法, 其特征在于, 所述计算调配物理资源数包 括:
计算调配物理资源数, 使得按照调配物理资源数调整后的每一个 CP物理 资源上为当前 CP的总工作负荷分摊的负荷值、 与调整后每一个 UP物理资源 上为当前 UP的总工作负荷分摊的负荷值之间的差值的绝对值或比值在一个预 设的范围阈值内。
Figure imgf000028_0001
NumA xALoadwork _ NumB xBLoadwork
Num . + X NumR - X 其中, 所述 Nwrn^是当前为 CP 和 UP 中的稀缺面分配的物理资源数, ALoadwork是 CP和 UP中的稀缺面的工作负荷, NumB是当前为 CP和 UP中的 冗余面分配的物理资源数, ^^ ^是 ?和1;?中的冗余面的工作负荷, 所 述 是调配物理资源数。
7、 如权利要求 1-6任一项所述的方法, 其特征在于, CP工作负荷和 UP 作负荷满足预设的均衡调整条件是指:
CP工作负荷和 UP工作负荷的差值的绝对值或比值在预设的范围阈值之
8、 如权利要求 1-7任一项所述的方法, 其特征在于, 所述根据计算得到 的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 将 CP的单板中所述调配物理资源数所指示数量的 单板的逻辑类型修改为 UP逻辑类型以实现 UP功能;
当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示数量的 单板的逻辑类型修改为 CP逻辑类型以实现 CP功能。
9、 如权利要求 1-7任一项所述的方法, 其特征在于, 所述根据计算得到 的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资 源数对应数量的 CP软件实例, 并在空闲出的物理资源上新建所述调配物理资 源数对应数量的 UP软件实例;
当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资 源数对应数量的 UP的软件实例, 并在空闲出的物理资源上新建所述调配物理 资源数对应数量的 CP软件实例。
10、 如权利要求 1-7任一项所述的方法, 其特征在于, 所述根据计算得到 的调配物理资源数调整 CP和 UP的物理资源, 包括:
当 CP作为冗余面时, 按照调配物理资源数的指示减少 CP的软件实例使 用的物理资源数, 并将空闲出的调配物理资源数对应数量的物理资源划拨给
UP软件实例使用;
当 UP作为冗余面时, 按照调配物理资源数的指示减少 UP软件实例使用 的物理资源数, 并将空闲出的调配物理资源数对应数量的物理资源划拨给 CP 软件实例使用。
11、 如权利要求 1-10任一项所述的方法, 其特征在于, 所述根据计算得 到的调配物理资源数调整 CP和 UP的物理资源之前, 还包括:
根据计算得到的调配物理资源数确定待调整的物理资源;
禁止向确定的待调整的物理资源上分配新业务和等待已经部署在确定的 待调整物理资源上的业务终止。
12、 如权利要求 1-11任一项所述的方法, 其特征在于, 所述冗余面是指 CP和 UP中,将获取到的 CP工作负荷和 UP工作负荷进 行比较, 其中工作负荷较小的一面;
所述稀缺面是指 CP和 UP中,将获取到的 CP工作负荷和 UP工作负荷进 行比较, 其中工作负荷较大的一面。
13、 如权利要求 1-12任一项所述的方法, 其特征在于, 所述根据计算得 到的调配物理资源数调整 CP和 UP的物理资源, 包括:
将调配物理资源数提交给操作系统 OS由 OS自动调度; 或者, 根据调配 物理资源数进行虚拟化调整; 或者根据调配物理资源数进行云计算调整。
14、 一种物理资源调整装置, 其特征在于, 包括:
收集模块, 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息,并根据对应的负荷信息确定得到 CP工作负荷和 UP 工作负荷;
计算模块, 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件 时, 计算调配物理资源数;
调整模块, 用于根据计算得到的调配物理资源数调整 CP和 UP的物理资 源。
15、 如权利要求 14所述的装置, 其特征在于, 所述调整模块包括: 调整单元, 用于根据计算得到的调配物理资源数自动调整 CP和 UP的物 理资源;
或者包括:
指示单元, 用于根据计算得到的调配物理资源数发出调整指示信息, 以指 示用户按照计算得到的调配物理资源数调整 CP和 UP的物理资源。
16、 如权利要求 14或 15所述的装置, 其特征在于, 所述收集模块包括: 获取单元, 用于从控制面 CP单板收集 CP的负荷信息, 并从用户面 UP 单板收集 UP的负荷信息, 并确定对应的 CP面负荷量和 UP面负荷量, 所述 负荷信息包括 CPU使用率、 用户数、 吞吐率中的任一种或多种;
记录单元, 用于将预设时间周期内 CP负荷量中的最大值或者负荷量的平 均值作为 CP的工作负荷, 预设时间周期内 UP负荷量中的最大值或者负荷量 的平均值作为 UP的工作负荷。
17、 如权利要求 14至 16任一项所述的装置, 其特征在于, 所述计算模块 包括:
确定单元, 用于在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件 时, 确定当前为 CP分配的 CP物理资源数和为 UP分配的 UP物理资源数; 计算单元, 用于根据当前为 CP分配的 CP物理资源数和 CP工作负荷、 当前为 UP分配的 UP物理资源数和 UP工作负荷, 计算调配物理资源数。
18、 如权利要求 17所述的装置, 其特征在于,
所述计算单元具体用于计算调配物理资源数, 使得调整后的每一个 CP物 理资源上为当前 CP的总工作负荷分摊的负荷值与调整后每一个 UP物理资源 上为当前 UP的总工作负荷分摊的负荷值之间的差值的绝对值或比值在一个预 设的范围阈值内。
19、 如权利要求 17所述的装置, 其特征在于, 所述计算模块还包括: 条件计算单元, 用于计算 CP工作负荷和 UP工作负荷差值的绝对值或比 值;
判断单元,用于判断所述条件计算单元计算得到的结果是否在预设的范围 阈值之外, 若是, 则 CP工作负荷和 UP工作负荷满足预设的均衡调整条件, 通知所述确定单元。
20、 如权利要求 14-19任一项所述的装置, 其特征在于,
所述调整模块, 具体用于当 CP作为冗余面时, 将 CP的单板中所述调配 物理资源数所指示数量的单板的逻辑类型修改为 UP逻辑类型以实现 UP功能; 当 UP作为冗余面时, 将 UP的单板中所述调配物理资源数所指示数量的单板 的逻辑类型修改为 CP逻辑类型以实现 CP功能。
21、 如权利要求如权利要求 14-19任一项所述的装置, 其特征在于, 所述调整模块, 具体用于当 CP作为冗余面时, 更改软件配置, 根据软件 配置删除所述调配物理资源数对应数量的 CP软件实例, 并在空闲出的物理资 源上新建所述调配物理资源数对应数量的 UP软件实例;当 UP作为冗余面时, 更改软件配置, 根据软件配置删除所述调配物理资源数对应数量的 UP的软件 实例, 并在空闲出的物理资源上新建所述调配物理资源数对应数量的 CP软件 实例。
11、 如权利要求如权利要求 14-19任一项所述的装置, 其特征在于, 所述调整模块, 具体用于当 CP作为冗余面时, 按照调配物理资源数的指 示减少 CP的软件实例使用的物理资源数, 并将空闲出的调配物理资源数对应 数量的物理资源划拨给 UP软件实例使用; 当 UP作为冗余面时, 按照调配物 理资源数的指示减少 UP软件实例使用的物理资源数, 并将空闲出的调配物理 资源数对应数量的物理资源划拨给 CP软件实例使用。
23、 如权利要求 14-22任一项所述的装置, 其特征在于, 还包括: 确定模块, 用于根据计算得到的调配物理资源数确定待调整的物理资源; 控制模块,用于禁止向确定的待调整的物理资源上分配新业务和等待已经 部署在确定的待调整物理资源上的业务终止。
24、 一种控制器, 其特征在于, 包括控制单板、 CP单板和 UP单板, 其 中:
所述控制单板用于从所述 CP单板收集 CP的负荷信息, 从所述 UP单板 收集 UP的负荷信息, 并根据对应的负荷信息确定得到 CP工作负荷和 UP工 作负荷; 在 CP工作负荷和 UP工作负荷满足预设的均衡调整条件时, 计算调 配物理资源数; 根据计算得到的调配物理资源数调整 CP单板和 UP单板的物 理资源。 25、 如权利要求 24所述的控制器, 其特征在于, 所述控制单板具体包括 如权利要求 14-23任一项所述的物理资源调整装置。
PCT/CN2013/076602 2013-05-31 2013-05-31 一种物理资源调整方法、装置及控制器 WO2014190557A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
EP13885716.4A EP2996382B1 (en) 2013-05-31 2013-05-31 Physical resource adjustment method, device and controller
PCT/CN2013/076602 WO2014190557A1 (zh) 2013-05-31 2013-05-31 一种物理资源调整方法、装置及控制器
CN201380001444.0A CN103703830B (zh) 2013-05-31 2013-05-31 一种物理资源调整方法、装置及控制器
US14/951,629 US10142997B2 (en) 2013-05-31 2015-11-25 Method and apparatus for adjusting physical resource, and controller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2013/076602 WO2014190557A1 (zh) 2013-05-31 2013-05-31 一种物理资源调整方法、装置及控制器

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US14/951,629 Continuation US10142997B2 (en) 2013-05-31 2015-11-25 Method and apparatus for adjusting physical resource, and controller

Publications (1)

Publication Number Publication Date
WO2014190557A1 true WO2014190557A1 (zh) 2014-12-04

Family

ID=50363950

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2013/076602 WO2014190557A1 (zh) 2013-05-31 2013-05-31 一种物理资源调整方法、装置及控制器

Country Status (4)

Country Link
US (1) US10142997B2 (zh)
EP (1) EP2996382B1 (zh)
CN (1) CN103703830B (zh)
WO (1) WO2014190557A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107959705B (zh) * 2016-10-18 2021-08-20 阿里巴巴集团控股有限公司 流式计算任务的分配方法和控制服务器
US10986516B2 (en) * 2017-03-10 2021-04-20 Huawei Technologies Co., Ltd. System and method of network policy optimization
CN107707424A (zh) * 2017-09-11 2018-02-16 厦门集微科技有限公司 负载状态的控制方法及系统
US11678252B2 (en) 2018-10-05 2023-06-13 Huawei Technologies Co., Ltd. Quality of service information notification to user equipment, users, and application server
CN111190728B (zh) * 2019-12-13 2023-08-25 北京山石网科信息技术有限公司 资源调整方法及装置
WO2022018466A1 (en) * 2020-07-22 2022-01-27 Citrix Systems, Inc. Determining server utilization using upper bound values

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101114927A (zh) * 2006-07-24 2008-01-30 华为技术有限公司 一种实现负载均衡的系统及方法
CN101447939A (zh) * 2008-12-16 2009-06-03 中国移动通信集团北京有限公司 一种业务分配方法和负载均衡器
EP2375658A1 (en) * 2010-04-07 2011-10-12 Vodafone Group PLC Traffic congestion in radio network controllers

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
FI20085217A0 (fi) * 2008-03-07 2008-03-07 Nokia Corp Tietojenkäsittelyjärjestely
US8681622B2 (en) * 2010-12-17 2014-03-25 Tekelec, Inc. Policy and charging rules function (PCRF) and performance intelligence center (PIC) based congestion control
US8516493B2 (en) * 2011-02-01 2013-08-20 Futurewei Technologies, Inc. System and method for massively multi-core computing systems
EP2592863B1 (en) * 2011-11-14 2014-01-08 Alcatel Lucent Distributed load balancing in a radio access network
CN103634801B (zh) * 2012-08-20 2018-02-09 中兴通讯股份有限公司 资源分配方法和装置
EP2918096B1 (en) * 2012-11-08 2020-01-08 Telefonaktiebolaget LM Ericsson (publ) Pcrf apparatus and traffic handling method for a pcrf

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101114927A (zh) * 2006-07-24 2008-01-30 华为技术有限公司 一种实现负载均衡的系统及方法
CN101447939A (zh) * 2008-12-16 2009-06-03 中国移动通信集团北京有限公司 一种业务分配方法和负载均衡器
EP2375658A1 (en) * 2010-04-07 2011-10-12 Vodafone Group PLC Traffic congestion in radio network controllers

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP2996382A4 *

Also Published As

Publication number Publication date
CN103703830B (zh) 2017-11-17
EP2996382A4 (en) 2016-04-20
EP2996382B1 (en) 2020-07-08
US20160081103A1 (en) 2016-03-17
US10142997B2 (en) 2018-11-27
EP2996382A1 (en) 2016-03-16
CN103703830A (zh) 2014-04-02

Similar Documents

Publication Publication Date Title
WO2014190557A1 (zh) 一种物理资源调整方法、装置及控制器
CN112867050B (zh) 一种upf网元管理方法及系统
CN103699445B (zh) 一种任务调度方法、装置及系统
JP6885403B2 (ja) リソース管理装置、リソース管理方法及びプログラム
WO2018072687A1 (zh) 一种资源调度的方法、装置和过滤式调度器
WO2014101010A1 (zh) 一种虚拟机系统的资源管理方法、虚拟机系统和装置
WO2014029308A1 (zh) 资源分配方法和装置
JP2016086225A (ja) ネットワークシステム、ゲートウェイ装置用プログラム及び管理装置用プログラム
CN112703774A (zh) 网络切片管理
WO2010124516A1 (zh) 数据通信输入端口调度方法及装置
WO2015014117A1 (zh) 云服务器及其虚拟机管理方法
WO2021136100A1 (zh) Cdn调度方法、系统、设备和存储介质
WO2011140866A1 (zh) 呼叫处理方法、装置和系统
WO2013082742A1 (zh) 资源调度方法、装置和系统
WO2018233679A1 (zh) 负荷均衡方法及装置
WO2014110743A1 (zh) 一种数据中心间资源调度方法和设备
WO2015184612A1 (zh) 一种资源调度方法,及装置
WO2016115846A1 (zh) 业务调度方法、装置、无线网络控制器及基站
WO2024051236A1 (zh) 资源调度方法及其相关设备
CN114064226A (zh) 容器集群的资源调协方法、调协装置及存储介质
WO2016197673A1 (zh) 一种载波聚合场景下的成员载波调度方法及装置
CN112347013A (zh) 一种中断处理方法以及相关装置
JP5829230B2 (ja) 管理システム及び管理方法
JP5997659B2 (ja) 分散処理システムおよび分散処理方法
CN108882296A (zh) 一种处理报文的方法及装置

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13885716

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

WWE Wipo information: entry into national phase

Ref document number: 2013885716

Country of ref document: EP