WO2016038857A1 - スケール数推定装置、スケール数管理システム、スケール数推定方法、スケール数管理方法、および、記憶媒体 - Google Patents
スケール数推定装置、スケール数管理システム、スケール数推定方法、スケール数管理方法、および、記憶媒体 Download PDFInfo
- Publication number
- WO2016038857A1 WO2016038857A1 PCT/JP2015/004477 JP2015004477W WO2016038857A1 WO 2016038857 A1 WO2016038857 A1 WO 2016038857A1 JP 2015004477 W JP2015004477 W JP 2015004477W WO 2016038857 A1 WO2016038857 A1 WO 2016038857A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- node
- service
- function
- rule
- scales
- Prior art date
Links
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/14—Network analysis or design
- H04L41/145—Network analysis or design involving simulating, designing, planning or modelling of a network
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/46—Multiprogramming arrangements
- G06F9/50—Allocation of resources, e.g. of the central processing unit [CPU]
- G06F9/5005—Allocation of resources, e.g. of the central processing unit [CPU] to service a request
- G06F9/5027—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals
- G06F9/505—Allocation of resources, e.g. of the central processing unit [CPU] to service a request the resource being a machine, e.g. CPUs, Servers, Terminals considering the load
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/40—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using virtualisation of network functions or resources, e.g. SDN or NFV entities
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/08—Configuration management of networks or network elements
- H04L41/0896—Bandwidth or capacity management, i.e. automatically increasing or decreasing capacities
Definitions
- the present invention relates to a technique for managing the number of scales of a node that provides a network function.
- the IT service provider uses a network function providing apparatus including various nodes that provide a network function.
- Nodes of the network function providing device include a load balancer (LB), a firewall (FW), and NAT (Network Address Translation).
- the traffic volume of IT services always fluctuates due to multiple factors such as the number of users and time zones.
- the IT service provider since various nodes that provide network functions are dedicated devices, it is difficult to control the network function providing apparatus and the throughput performance of the nodes. Therefore, the IT service provider has to adjust the amount of traffic to be processed by the network function providing apparatus in accordance with the throughput performance on the network function providing apparatus side.
- NFV Network Function Virtualization
- SDN Software Defined Networking
- nodes such as FW and LB are realized by software.
- VNF Virtualized Network Function
- VNFC Virtualized Network Function
- the processing performance can be controlled by scaling the VNFC.
- Each VNFC is a separate virtual machine.
- the VNFC adjusts network functions by setting function rules for providing functions according to network requirements. For example, in the case of VNFC that provides a firewall function, a function rule as shown in FIG. 18 is set.
- a VNFC in which such functional rules are set can provide functions such as http (Hypertext transfer protocol) access, ftp (File transfer protocol) access permission, and attack avoidance.
- Patent Document 1 An example of related technology for managing the performance of a network function providing apparatus using such network function virtualization technology is described in Patent Document 1.
- resources for each agent for example, CPU: Central Processing Unit, RAM: Random Access Memory, etc.
- Patent Document 1 has the following problems.
- the setting contents of the node functional rules affect the processing performance of node instances such as the above-mentioned VNFC.
- a node instance consumes more resources such as a CPU as the setting contents of function rules to be set increases, and the time required for processing becomes longer. Therefore, the processing performance of the node instance changes according to the setting content of the function rule.
- Patent Document 1 does not describe estimating the processing performance of the agent and estimating the number of parallel processes (scale number) of the agent. Therefore, in this related technique, the performance of the agent cannot be sufficiently controlled only by reallocating resources when the performance does not satisfy the target value.
- an object of the present invention is to provide a technique for more accurately estimating the processing performance of an instance in a node that provides a network function and more accurately estimating the number of scales that can process input traffic.
- the scale number estimation apparatus of the present invention calculates a function rule quantitative value obtained by quantifying a function rule based on a function rule setting history set in a node that provides a network function.
- a rule quantification unit a service amount calculation unit that calculates a service amount per unit time of a node instance operating in the node based on a service history of the node; and the functional rule quantification value obtained for the node;
- a performance model generating unit that generates a performance model representing a relationship between the amount of input traffic to the node, the function rule quantitative value, and the number of node instances (number of scales);
- the expected input traffic volume and the functional rule definition Comprising a number of scales estimating means for estimating the number of scales in accordance with the value, the.
- the scale number management system of the present invention includes a network function providing device including a node that provides a network function, a function rule setting history (function rule setting history) set in the node from the network function providing device, and The above-mentioned scale for estimating the number of scales of the node in the network function providing device using the acquisition device for acquiring the service history of the node, and the function rule setting history and the service history acquired by the acquisition device A number estimation device; and a control device that controls the number of scales of nodes in the network function providing device based on the number of scales estimated by the scale number estimation device.
- a network function providing device including a node that provides a network function, a function rule setting history (function rule setting history) set in the node from the network function providing device, and The above-mentioned scale for estimating the number of scales of the node in the network function providing device using the acquisition device for acquiring the service history of the node, and the function rule setting history and the service history acquired by the acquisition device A number
- the scale number estimation method of the present invention calculates a function rule quantitative value obtained by quantifying the function rule based on the function rule setting history set in the node that provides the network function, and stores the function rule in the service history of the node. Based on the set of the functional rule quantitative value and the service amount obtained for the node, based on the set of the function rule quantitative value and the service amount obtained for the node, A performance model representing the relationship between the functional rule quantitative value and the number of node instances (number of scales) is generated, and the performance model is used to respond to the assumed input traffic amount and the functional rule quantitative value The number of scales is estimated.
- the scale number management method of the present invention obtains a function rule setting history (function rule setting history) set in the node in a network function providing device including a node that provides a network function, and the network function providing device.
- a service history of the node in the network and based on the acquired function rule setting history and service history, using the scale number estimation method described above, the number of scales of the node in the network function providing device is estimated, the estimated scale Based on the number, the scale number of the node in the network function providing device is controlled.
- the storage medium of the present invention includes a function rule quantification step of calculating a function rule quantification value obtained by quantifying a function rule based on a function rule setting history set in a node that provides a network function;
- a service amount calculation step of calculating a service amount per unit time of a node instance operating in the node based on the service history of the node, and based on the set of the functional rule quantitative value and the service amount obtained for the node
- a performance model generation step for generating a performance model representing a relationship between the amount of input traffic to the node, the quantitative value of the function rule, and the number of node instances (number of scales), and the performance model is assumed.
- the traffic according to the input traffic volume and the function rule quantitative value It stores a computer program for executing the scale number estimating step of estimating the number Lumpur, to the computer device.
- the present invention can provide a technique for more accurately estimating the processing performance of an instance in a node that provides a network function, and more accurately estimating the number of scales that can process input traffic.
- the scale number management system 1 includes a scale number estimation device 10, a network function providing device 30, an acquisition device 50, and a control device 70.
- the scale number estimation device 10 is connected to the acquisition device 50 and the control device 70 so as to communicate with each other.
- the network function providing device 30 is connected to the acquisition device 50 and the control device 70 so that they can communicate with each other.
- the scale number estimation apparatus 10 includes a function rule quantification unit 11, a service amount calculation unit 12, a performance model generation unit 13, a performance model storage unit 130, and a scale number estimation unit 14.
- the performance model storage unit 130 constitutes an embodiment of a part of the performance model generation unit of the present invention.
- the network function providing device 30 includes one or more nodes 31. In the node 31, one or more node instances 32 operate.
- the acquisition device 50 includes a function rule setting history storage unit 501 and a service history storage unit 502.
- each of the scale number estimation device 10, the network function providing device 30, the acquisition device 50, and the control device 70 may be configured by hardware elements as shown in FIG.
- the scale number estimation device 10 can be configured by the computer device 100.
- the computer apparatus 100 includes a CPU (Central Processing Unit) 1001, a RAM (Random Access Memory) 1002, a ROM (Read Only Memory) 1003, a storage device 1004 such as a hard disk, and a network interface 1005.
- the network function providing device 30 can be configured by the computer device 300.
- the computer apparatus 300 includes a CPU 3001, a RAM 3002, a ROM 3003, a storage device 3004 such as a hard disk, and a network interface 3005.
- the acquisition device 50 can be configured by the computer device 500.
- the computer device 500 includes a CPU 5001, a RAM 5002, a ROM 5003, a storage device 5004 such as a hard disk, and a network interface 5005.
- the control device 70 can be configured by a computer device 700.
- the computer device 700 includes a CPU 7001, a RAM 7002, a ROM 7003, a storage device 7004 such as a hard disk, and a network interface 7005.
- the scale number estimation device 10 the acquisition device 50, and the control device 70 are communicably connected by network interfaces 1005, 5005, and 7005, respectively.
- the network function providing device 30, the acquisition device 50, and the control device 70 are communicably connected by network interfaces 3005, 5005, and 7005, respectively.
- the network function providing apparatus 30 includes a network interface (not shown) for connection to a service side or a terminal side as a network function providing destination.
- each of these functional blocks includes a network interface 1005 and a CPU 1001 that reads a computer program and various data stored in the ROM 1003 and the storage device 1004 into the RAM 1002 and executes them.
- the performance model generation unit 13 includes a CPU 1001 that reads a computer program and various data stored in the ROM 1003 and the storage device 1004 into the RAM 1002 and executes them.
- the node 31 and the node instance 32 are configured by a CPU 3001 that reads a computer program and various data stored in the ROM 3003 and the storage device 3004 into the RAM 3002 and executes them. Further, the function rule setting history storage unit 501 and the service history storage unit 502 of the acquisition device 50 are configured by a storage device 5004. Note that the hardware configuration of each device and each functional block thereof is not limited to the above-described configuration. For example, some or all of the devices may be realized on the same computer device. As a specific example, the scale number estimation device 10 and the acquisition device 50 may be realized on the same computer device, and the network function providing device 30 and the control device 70 may be realized on the same computer device. In this case, devices that require information transmission / reception may be connected by inputting / outputting information via a storage device instead of being connected via a network interface.
- the node 31 of the network function providing device 30 provides a network function based on the set function rule.
- the node instance 32 is an instance of the node 31, and each is realized as a virtual machine.
- the operating node instance 32 executes a process (service) for providing a network function in response to a request from the outside.
- the acquisition device 50 acquires the function rule setting history of each node 31 and the service history of each node 31 from the network function providing device 30.
- the acquisition device 50 stores the acquired function rule setting history in the function rule setting history storage unit 501.
- the acquisition device 50 stores the acquired service history in the service history storage unit 502.
- the function rule setting history represents a function rule setting history set in the node 31.
- the function rule setting history may be a setting information history including the setting contents of the function rule set in the node 31 and the setting time.
- the acquisition device 50 further acquires a function rule setting history when a function rule is set for a new node 31 in the network function providing device 30 or when a function rule of an existing node 31 is changed. Then, the acquisition device 50 adds and stores the newly acquired function rule setting history in the function rule setting history storage unit 501.
- the service history may be information including a set of a service time required for processing executed by the node instance 32 of each node 31 and a processed data amount. Further, when a new process is executed by the node instance 32 in the network function providing device 30, the acquisition device 50 further acquires the service history, adds it to the service history storage unit 502, and stores it.
- the function rule quantification unit 11 of the scale number estimation device 10 calculates a function rule quantification value obtained by quantifying the function rule based on the function rule setting history of the node 31.
- the function rule fixed value is information that can quantitatively express the setting contents of the function rule.
- the scale number estimation device 10 acquires a function rule setting history from the function rule setting history storage unit 501 of the acquisition device 50.
- the function rule quantification unit 11 calculates a function rule quantification value for each function rule setting information for each node 31. If there are a plurality of setting information for the same node 31, the function rule quantification unit 11 calculates a plurality of function rule quantification values for the same node 31.
- the service amount calculation unit 12 calculates the service amount per unit time of the node instance 32 based on the service history of the node 31. Specifically, the service amount calculation unit 12 acquires a service history from the acquisition device 50. For example, the service amount calculation unit 12 can calculate the service amount per unit time based on the service time and the data amount included in the service history.
- the service amount calculation unit 12 stores the service amount calculated for each node 31 in the storage device 1004 in association with the function rule quantitative value of the node 31. If a plurality of functional rule quantitative values are calculated for one node 31, the service amount calculation unit 12 for each functional rule quantitative value is based on the service history during the period in which the functional rule quantitative value is valid. The service amount described above may be obtained. Then, the service amount calculation unit 12 associates the calculated service amount with the corresponding function rule quantitative value.
- the effective period of the function rule fixed value can be calculated from the function rule setting time included in the function rule setting history.
- the performance model generation unit 13 Based on the set of functional rule quantitative values and service amounts obtained for each node 31, the performance model generation unit 13 inputs the amount of traffic input to the node 31, the functional rule quantitative value, and the number of node instances 32 (number of scales). Generate a performance model that represents the relationship between That is, the performance model includes a calculation formula capable of calculating the number of node instances 32 according to the function rule quantitative value of the node 31 and the input traffic amount. The performance model generation unit 13 stores the performance model generated for each node 31 in the performance model storage unit 130.
- the scale number estimation unit 14 estimates the scale number using the performance model of the node 31 based on the input traffic amount assumed for the node 31 and the function rule quantitative value. In addition, the scale number estimation unit 14 outputs information indicating the scale number estimated for each node 31 to the control device 70.
- the control device 70 controls the number of node instances 32 in the node 31 in the network function providing device 30 based on the scale number of each node 31 output from the scale number estimation device 10.
- the acquisition device 50 acquires the function rule setting history of the node 31 from the network function providing device 30 (step A1). As described above, the acquisition device 50 stores the acquired function rule setting history in the function rule setting history storage unit 501.
- the service amount calculation unit 12 acquires the service history of the node 31 from the network function providing device 30 (step A2). As described above, the acquisition device 50 stores the acquired service history in the service history storage unit 502.
- step A1 and step A2 need not be executed in this order. Further, the operations of step A1 and step A2 may be executed substantially simultaneously. Further, the operations of Step A1 and Step A2 may be repeatedly executed for a specified period.
- the function rule quantification unit 11 reads out the function rule setting history and the service history from the function rule setting history storage unit 501 and the service history storage unit 502 of the acquisition device 50 (step B1).
- the function rule quantification unit 11 repeats the following steps B2 to B4 for each node 31 recorded in the function rule setting history.
- the function rule quantification unit 11 calculates a function rule quantification value from the function rule setting history for this node 31 (step B2).
- the function rule quantification unit 11 may obtain a function rule quantification value for each setting information about the node 31.
- the service amount calculation unit 12 calculates the service amount per unit time of the node instance 32 operating in this node 31 for each function rule fixed value of this node 31 obtained in step B2 (step B3). .
- the service amount calculation unit 12 may obtain the corresponding service amount using the service history of this node 31 during the validity period of each functional rule quantitative value.
- the service amount calculation unit 12 may obtain the service amount based on the data amount and service time of each process included in the corresponding service history. Then, the service amount calculation unit 12 associates a corresponding service amount with each function rule fixed value.
- the performance model generation unit 13 based on the set of functional rule quantitative values and service amount sets obtained for the node 31, the input traffic amount, the functional rule quantitative value, the scale number, A performance model representing the relationship is generated (step B4). Then, the performance model generation unit 13 stores the performance model generated for the node 31 in the performance model storage unit 130.
- the scale number estimation unit 14 acquires the input traffic amount assumed for the estimation target node 31 and the contents of the functional rules set in the node 31 (step C1). For example, the scale number estimation unit 14 acquires the amount of input traffic assumed for the node 31 to be estimated and the content of the function rule from the input device (not shown), the network interface 1005, the storage device 1004, or the like. May be.
- the function rule quantification unit 11 calculates a function rule quantification value based on the function rule acquired in step C1 (step C2).
- the scale number estimation unit 14 acquires the performance model of the node 31 from the performance model storage unit 130. Then, the scale number estimating unit 14 applies the input traffic amount acquired in Step C1 and the function rule quantitative value calculated in Step C2 to the performance model of the node 31. Thereby, the scale number estimation part 14 calculates the scale number of this node 31, and outputs it to the control apparatus 70 (step C3).
- control device 70 acquires the number of scales output from the scale number estimation device 10 (step D1).
- control device 70 controls the number of node instances 32 in the node 31 on the network function providing device 30 based on the acquired scale number (step D2). For example, if the number of active node instances 32 for the node 31 is different from the estimated number of scales, the control device 70 changes the number of node instances 32 so as to be the estimated number of scales. If this node 31 is not yet operating, node instances 32 of this node 31 are generated and operated for the estimated number of scales.
- control device 70 ends the scale number control operation.
- the scale number management system 1 may repeat the operations of steps C1 to C3 and D1 to D2 described above for each node 31 that is operating or scheduled to operate in the network function providing device 30.
- the scale number management system estimates the processing performance of an instance in a node providing a network function with higher accuracy, and controls the number of scales capable of processing input traffic with higher accuracy. Can do.
- the acquisition device acquires the function rule setting history and service history set in each node of the network function providing device.
- the function rule quantification unit of the scale number estimation device calculates a function rule quantification value based on the function rule setting history.
- the service amount calculation unit calculates the service amount per unit time of the node instance of each node based on the service history.
- generation part produces
- the scale number estimation unit estimates the number of node instances corresponding to the assumed input traffic volume using the performance model. This is because the control device controls the number of node instances based on the estimated number of scales.
- the present embodiment can improve the estimation accuracy of the processing performance of the instance in the node. As a result, the present embodiment can more accurately estimate the number of scales that can process the expected input traffic with the minimum necessary resources in accordance with the contents of the function rules set in the node and changes thereof. .
- the service chain execution device is a device that executes a service chain that connects and functions a plurality of nodes. Note that, in each drawing referred to in the description of the present embodiment, the same reference numerals are given to the same configuration and steps that operate in the same manner as in the first embodiment of the present invention, and the detailed description in the present embodiment. Description is omitted.
- FIG. 7 shows the configuration of a scale number management system 2 as a second embodiment of the present invention.
- the scale number management system 2 includes a scale number estimation device 20, a service chain execution device 40, an acquisition device 50, and a control device 80.
- the service chain execution device 40 constitutes an embodiment of the network function providing device of the present invention.
- the scale number estimation device 20 and the service chain execution device 40 are communicably connected to the acquisition device 50 and the control device 80, respectively.
- the scale number estimation apparatus 20 includes a function rule quantification unit 21, a service amount calculation unit 22, a performance model generation unit 23, a performance model storage unit 230, and a chain instance generation unit 24.
- the performance model storage unit 230 constitutes an embodiment of the performance model generation unit of the present invention.
- the chain instance generation unit 24 constitutes an embodiment of the scale number estimation unit of the present invention.
- the service chain execution device 40 includes a service chain 43, a node 31, and a node instance 32.
- each device and each functional block constituting the scale number management system 2 can be configured by the hardware elements shown in FIG. 2 as in the first embodiment of the present invention. Note that the hardware configuration of each device and each functional block thereof is not limited to the above-described configuration.
- the service chain execution device 40 is a device that executes a service chain 43 that connects and functions a plurality of nodes 31.
- the service chain 43 provides a series of network functions corresponding to the contents of the IT service in cooperation with a device that provides various IT services for the terminal.
- a service chain 43 that provides a network function necessary for providing a web service to a terminal, a service chain 43 that is necessary for providing a moving image distribution system, and the like.
- the service chain 43 is defined by definition information.
- the definition information of the service chain 43 may be information including each node 31 to be connected and a function rule set in each node 31.
- the service chain 43 is configured to operate by generating an instance (chain instance) of the service chain 43 based on the definition information of the service chain 43.
- the service chain 43 in operation provides a series of network functions by sequentially processing incoming incoming traffic by the node instances 32 of each node 31.
- one or more service chains 43 can be operated.
- the acquisition device 50 is configured similarly to the first embodiment of the present invention. Thereby, the acquisition device 50 acquires the function rule setting history and service history of each node 31 constituting the service chain 43 from the service chain execution device 40.
- the function rule setting history includes a node ID for identifying the node 31, information indicating the contents of the function rule, and information (time stamp) indicating when the setting of the function rule is valid. Shall be included.
- the service history includes a node ID, a process start time and an end time, and a processed data amount.
- the acquisition device 50 stores such function rule setting history and service history in the function rule setting history storage unit 501 and the service history storage unit 502.
- the function rule quantification unit 21 of the scale number estimation device 20 refers to the function rule setting history acquired from the service chain execution device 40 and calculates a function rule quantification value based on the number of function rules.
- the function rule quantitative value may be the number of function rules itself.
- the function rule quantification unit 21 associates the node ID and the time stamp with the calculated function rule quantification value.
- the service amount calculation unit 22 refers to the service history acquired from the service chain execution device 40 and calculates the service amount for each node 31 for each valid period of the function rule fixed value. Then, the service amount calculation unit 22 associates the service amount calculated for each effective period with the function rule quantitative value in the effective period and the corresponding node ID.
- the service history includes the node ID, the processing start time and end time, and the amount of data processed.
- the function rule quantifying unit 21 associates the node ID, the function rule quantified value, and the time stamp. Therefore, the service amount calculation unit 22 first calculates an effective period for each function rule fixed value calculated for the node 31 having the same node ID. Specifically, the service amount calculation unit 22 from the time stamp of a function rule fixed value with a certain node ID to the next newest time stamp among the time stamps of other function rule fixed values with the same node ID. May be the valid period.
- the service amount calculation unit 22 calculates the average value of the data amount of the service time from the start time to the end time in the service history of the effective period including the corresponding node ID.
- the service amount may be obtained by dividing by the average value.
- the service history of the valid period may be a service history in which both or one of the start time and the end time is included in the valid period. Then, the service amount calculation unit 22 associates the node ID, the function rule fixed value, and the service amount.
- the function rule fixed value r1 and the function rule fixed value r2 are obtained for a certain node 31.
- the time stamp t1 is associated with the function rule fixed value r1
- the time stamp t2 is associated with the function rule fixed value r2.
- the time stamp t2 is newer than the time stamp t1.
- the service amount calculation unit 22 calculates time stamps t1 to t2 as the valid period of the function rule fixed value r1.
- the service amount calculation unit 22 calculates the current time from the time stamp t2 as the valid period of the function rule fixed value r2.
- the current time point here may be, for example, the time of processing for calculating the effective period, or may be up to the latest time point when the service history is obtained.
- the service amount calculation unit 22 calculates the service amount s1 for the node 31 based on the service history from the time stamps t1 to t2, and associates the node ID with the function rule fixed value r1. Further, the service amount calculation unit 22 calculates the service amount s2 for the same node 31 based on the service history from the time stamp t2 to the present time, and associates the node ID with the function rule fixed value r2.
- the performance model generation unit 23 generates, as a performance model, a service amount estimation formula and an equation representing the relationship between the service amount, the input traffic amount, and the number of scales.
- the performance model generation unit 23 uses the set of the service amount and the function rule quantitative value obtained for each node 31 for each valid period of the function rule quantitative value to set the service amount as an objective variable, function Statistical analysis using rule quantitative values as explanatory variables. Thereby, the performance model production
- the service amount estimation formula is represented by the following formula (1), for example. ... (1)
- rule n1 indicates the function rule quantitative value of the node n1
- and ⁇ n1 indicates the service amount of the node n1.
- the service amount can be estimated from the function rule quantitative value of a certain node 31.
- the performance model generation unit 23 employs the following equation (2) used in the queue model (M / M / S) having a plurality of windows as a model representing the behavior of the node 31.
- ⁇ ⁇ / S ⁇ ...
- ⁇ indicates the amount of traffic arriving at the node 31 (for example, Mbps: megabytes per second).
- ⁇ represents the service amount of the node 31 estimated by the above-described service amount estimation formula (1).
- S indicates the number of scales (the number of node instances 32), which is the number of parallel processes in the node 31.
- ⁇ represents an operation rate and represents the degree of processing congestion (0 to 1) in each node 31.
- the operating rate ⁇ represents that the waiting time is longer as the operating rate is closer to 1.
- a specified value is set for the operating rate ⁇ .
- the performance model generation unit 23 generates a service amount estimation formula of Expression (1) and an equation of Expression (2) as a performance model. For example, the performance model generation unit 23 associates Equation (1), Equation (2), the value of the operation rate ⁇ , and the node ID, and stores them in the performance model storage unit 230.
- the chain instance generation unit 24 acquires the definition information of the service chain 43 and information on the amount of input traffic assumed in the service chain 43.
- the definition information of the service chain 43 includes each node 31 to be connected and the content of the function rule set in each node 31.
- the information on the assumed input traffic amount may be, for example, the maximum input traffic amount assumed as processing by the service chain 43.
- the chain instance generation unit 24 estimates the number of scales corresponding to the input traffic amount and the function rule quantitative value for each node 31 included in the definition information of the service chain 43 using the performance model. Specifically, for each node 31, the chain instance generation unit 24 applies the function rule quantitative value obtained from the definition information and the acquired input traffic amount to the equations (1) and (2) of the performance model. By doing so, the number of scales may be obtained.
- the chain instance generation unit 24 generates chain instance information of the service chain 43 using the estimated number of scales and the definition information of the service chain 43.
- the chain instance information is information necessary for generating a chain instance.
- the chain instance information may be information including the scale number of each included node 31 in addition to the information included in the definition information of the service chain 43.
- the chain instance generation unit 24 outputs the generated chain instance information to the control device 80.
- the control device 80 generates an instance of the service chain 43 based on the chain instance information. Specifically, the control device 80 may generate, on the service chain execution device 40, the node instances 32 of the set scale number for each node 31 included in the chain instance information according to the function rules. If the chain instance indicated by the chain instance information is already running on the service chain execution device 40, the control device 80 may adjust the scale number of each node 31 to the scale number included in the chain instance information.
- the operation in which the acquisition device 50 acquires the function rule setting history and the service history from the service chain execution device 40 is the same as the history acquisition operation in the first embodiment of the present invention described with reference to FIG. .
- information including the node ID, the function rule setting contents, and the time stamp is acquired as the function rule setting history.
- the service history information including a node ID, a process start time and an end time, and a data amount is acquired.
- the function rule quantification unit 21 executes step B1 as in the first embodiment of the present invention, and reads the function rule setting history and the service history.
- the function rule quantification unit 21 repeats the following steps B12 to B14 for each node 31 recorded in the function rule setting history.
- the function rule quantification unit 21 counts the number of function rules of each setting information included in the function rule setting history related to the node 31 to obtain a function rule quantified value. Then, the function rule quantification unit 21 associates the node ID of the node 31, the calculated function rule quantification value, and the time stamp included in the corresponding setting information (step B12).
- the service amount calculation unit 22 repeats the following steps B13 to B14 for each function rule quantitative value of the node 31 obtained in step B12.
- the service amount calculation unit 22 calculates the effective period of this functional rule quantitative value (step B13).
- the service amount calculation unit 22 may calculate the effective period from the time stamp of the setting information corresponding to this function rule quantitative value to the time stamp of the next new setting information for this node 31.
- the service amount calculation unit 22 calculates, as the service amount, a value obtained by dividing the average value of the data amount of each process by the average value of the service time in the service history of the calculated effective period. Then, the service amount calculation unit 22 associates the function rule fixed value with the calculated service amount and the node ID (step B14).
- the performance model generation unit 23 next generates a performance model for this node 31.
- the performance model generation unit 23 generates the service amount estimation formula (1) using the set of the obtained functional rule quantitative value and service amount (step B15).
- the performance model generation unit 23 may determine A and B in the above-described service amount estimation formula (1) by performing statistical analysis using the service amount as an objective variable and the rule amount as an explanatory variable. .
- the performance model generation unit 23 generates a performance model including the service amount estimation formula (1) generated in step B15 and the equation (2) used in the queue model (M / M / S) of multiple windows. Generate (step B16).
- the performance model generation unit 23 calculates the service amount estimation formula (1), the equation (2) used in the queue model (M / M / S) of multiple windows, the specified value of the operation rate ⁇ , May be stored in the performance model storage unit 230 in association with the node ID of the node 31.
- the scale number management system 2 ends the performance model generation operation.
- the chain instance generation unit 24 acquires the definition information of the service chain 43 and the amount of input traffic assumed to flow into the service chain 43 (step C11).
- the chain instance generation unit 24 may acquire such definition information and input traffic volume from an input device (not shown), the network interface 1005, the storage device 1004, or the like.
- the definition information of the service chain 43 includes the node ID of the node 31 to be linked in each service chain 43 and the function rule set in each node 31. Further, as the assumed input traffic volume, the maximum input traffic volume that each service chain 43 is supposed to process may be acquired.
- the chain instance generation unit 24 repeats the operations of the following steps C12 to C15 for each node 31 that is a constituent element thereof.
- the chain instance generation unit 24 extracts the performance model of the node 31 from the performance model storage unit 230 (step C12).
- the chain instance generation unit 24 calculates a function rule quantitative value from the function rule of the node 31 included in the service chain definition information using the function rule quantification unit 21 (step C13).
- the chain instance generation unit 24 substitutes the function rule quantitative value calculated in step C13 for the service amount estimation formula (1) of the performance model of the node 31. Thereby, the chain instance generation unit 24 calculates the service amount (step C14).
- the chain instance generation unit 24 substitutes the service amount calculated in Step C14 and the input traffic amount acquired in Step C11 into Equation (2) included in the performance model of the node 31. Thereby, the chain instance production
- the chain instance generation unit 24 When the processing of steps C12 to C15 is completed for each node 31 constituting this service chain 43, the chain instance generation unit 24 generates chain instance information and outputs it to the control device 80 (step C16).
- the chain instance information represents information for generating an instance of the service chain 43.
- the chain instance generation unit 24 may generate chain instance information based on the definition information of the service chain 43 and the scale number estimated for each node 31.
- the scale number estimation device 20 ends the chain instance generation operation.
- control device 80 acquires chain instance information from the scale number estimation device 20 (step D11).
- control device 80 repeats the following steps D12 to D14 for each chain instance included in the acquired chain instance information.
- the control device 80 when the corresponding chain instance has not yet been created (No in Step D12), the control device 80 generates this chain instance on the service chain execution device 40. Specifically, the control device 80 generates as many node instances 32 of the nodes 31 that constitute the chain instance as the number of scales (step D13).
- control device 80 changes the number of node instances 32 to be the scale number for each node 31 included in the chain instance (Ste D14).
- the scale number management system 2 ends the scale number control operation.
- nodeX a node ID for identifying each node 31 is expressed as “nodeX” or the like, and the node 31 whose node ID is nodeX is also simply referred to as “nodeX”.
- the acquisition device 50 acquires the function rule setting history for each node 31 in the service chain 43 from the service chain execution device 40 and stores it in the function rule setting history storage unit 501 (step A1).
- the acquired function rule setting history includes a node ID, function rule setting contents, and a time stamp.
- the function rule setting history shown in FIG. 11 has been acquired.
- the time stamp “2014/03/09: 09: 00: 00.000” includes two functions related to SSH (Secure SHell) and four functions related to DNS (Domain Name System). Rules are set.
- the function rule setting history shown in FIG. 11 includes information indicating the type of function of the node 31 in each setting information.
- the function rule setting history in this embodiment includes at least the node ID, the function rule, and the like.
- the setting contents and the time stamp may be included.
- the acquisition device 50 acquires the service history of each node 31 in the service chain 43 from the service chain execution device 40 and stores it in the service history storage unit 502 (step A2).
- the acquired service history includes a node ID, a process start time and an end time, and a processed data amount.
- the service history shown in FIG. 12 includes information indicating the function type of the node 31 in the history of each process.
- the service history in this embodiment includes at least the node ID, the start time of the process, It only needs to include the end time and the amount of data.
- the acquisition device 50 repeats the processing of steps A1 and A2 for a certain period, and accumulates the function rule setting history and service history.
- the performance model generation unit 23 counts the number of function rules for each setting information of each node 31 from the information of FIG. 11 stored in the function rule setting history storage unit 501, and determines the function rule quantitative value. To do. Then, the node ID, the function rule quantitative value, and the time stamp are associated (step B12). Thereby, the function rule fixed value information shown in FIG. 13 is generated.
- the functional rule quantitative value information shown in FIG. 13 includes information indicating the type of function of the node 31, but the functional rule quantitative value information in the present embodiment includes at least a node ID, a functional rule quantitative value, and And a time stamp.
- the service amount calculation unit 22 obtains a valid period from when the function rule is activated until it is updated based on the function rule quantitative value information (step B13). Then, the service amount calculation unit 22 obtains the service amount of the node 31 for each valid period of the function rule fixed value (step B14).
- the service amount calculation unit 22 obtains the service history of nodeX in the period from t1 to t2 from the service history of FIG. Then, the service amount ⁇ indicating the processing capability is calculated by dividing the average value of the data amount by the average value of the difference between the start time and the end time (service time) of each history. For example, when the unit of data amount is Mb (megabytes) and the unit of service time is s (seconds), the unit of service amount is “Mbps (megabytes per second)”.
- the service amount calculation unit 22 repeats such processing for each setting information of each node 31 to generate a set of sets of function rule quantitative values and service amount values for each node 31.
- This service amount estimation formula (1 ′) represents that the service amount decreases as the function rule quantitative value (number of rules) increases (step B15).
- the performance model generation unit 23 calculates, for each node 31, the service amount estimation formula (1) obtained, the equation (2) used in the queue model of multiple windows, and the operation rate ⁇ (here 0.7).
- the performance model storage unit 230 stores the performance model shown in FIG.
- the performance model shown in FIG. 14 includes information indicating the type of function of the node 31, but the performance model according to the present embodiment includes at least a node ID, a service amount estimation formula, a service amount, and an input traffic amount. It only needs to include an equation representing the relationship with the number of scales and the value of the operation rate.
- the chain instance generation unit 24 acquires service chain definition information shown in FIG. 15 as definition information related to the service chain 43 generated on the service chain execution device 40.
- the service chain 43 identified by the ID chain1 is configured to connect node1 functioning as FW, node3 functioning as NAT, and node7 functioning as LB to function.
- the service chain 43 identified by the ID chain2 is configured to connect node2 functioning as FW and node6 functioning as DPI (Deep Packet Inspection).
- the service chain definition information includes the node IDs constituting the service chain 43 and the setting contents of the function rules of each node 31.
- the chain instance generation unit 24 acquires information on the maximum expected input traffic volume for each service chain 43 included in the service chain definition information shown in FIG. For example, the acquired information is as shown in FIG. 16 (step C11).
- the chain instance generation unit 24 extracts the performance model of each node 31 that is a component of the service chain 43 from the information illustrated in FIG. 14 stored in the performance model storage unit 230. (Step C12).
- the chain instance generation unit 24 uses the function rule quantification unit 21 to quantify the function rules set in the definition information for each node 31 that is a component of each service chain 43. For example, in the service chain definition information of FIG. 15, two function rules are set for node1. Therefore, the function rule quantitative value of node1 is “2” (step C13).
- the chain instance generation unit 24 substitutes the function rule quantitative value of the node 31 obtained from the definition information into the service amount estimation formula (1) included in the performance model of the node 31.
- the chain instance generation unit 24 executes the processing of steps C12 to C15 for each node 31.
- the chain instance generation unit 24 generates chain instance information as shown in FIG. 17 by using the estimated scale number of each node 31 and the service chain definition information of FIG. 15 (step C16).
- the chain instance information shown in FIG. 17 includes the function rule quantitative value of each node 31, but it is not always necessary to include it.
- the chain instance information of the present embodiment only needs to include at least the scale number of each node 31 constituting each chain instance in addition to the service chain definition information.
- the control device 80 reads the chain instance information of FIG. 17 (step D11), and creates a chain instance on the service chain execution device 40. Specifically, the control device 80 generates and operates a node instance 32 having a designated scale number for each node 31 of each service chain 43. Alternatively, for the service chain 43 that is already in operation, the control device 80 adjusts the number of node instances 32 of each node 31 to be the scale number included in the chain instance information (steps D12 to D14).
- the scale number management system estimates the processing performance of the instance in the node of the service chain execution apparatus with higher accuracy, and controls the number of scales that can process the input traffic with higher accuracy. Can do.
- the acquisition device acquires the function rule setting history and service history of each node of the service chain.
- the function rule quantification unit of the scale number estimation device calculates a function rule quantification value based on the number of function rules of each node constituting the service chain.
- the service amount calculation unit calculates the service amount per unit time of the node instance by dividing the average value of the data amount processed in each node by the average value of the service time.
- generation part produces
- generation part produces
- the scale number estimation unit estimates the number of scales according to the input traffic amount assumed to flow into the service chain and the assumed function rule using the performance model. This is because the control device generates a service chain so that the number of node instances is based on the estimated number of scales.
- this embodiment can improve the estimation accuracy of the processing performance of the node instance according to the contents of the function rule set in each node of the service chain or the change thereof. As a result, the present embodiment can more accurately estimate the number of scales that can process the input traffic assumed to flow into the service chain with the minimum required resources.
- the function rule quantification unit calculates the function rule quantified value based on the number of function rules.
- the function rule quantification unit may calculate information that can be expressed by quantifying the setting content of the function rule as a function rule quantification value, or a value based on a combination of such information.
- the service amount calculation unit has been described mainly with respect to an example in which the service amount is calculated from the average value of the service time and the average value of the data amount of the node instances in each node.
- the service amount may be obtained for each node by the other calculation method using the service history of each node.
- the performance model may be another model as long as the number of scales can be estimated from the function rule quantitative value and the input traffic amount.
- each functional block of each device constituting the scale number management system is realized by a CPU that executes a computer program stored in a storage device or ROM.
- a storage device or ROM a storage device or ROM.
- some, all, or a combination of each functional block of each device may be realized by dedicated hardware.
- each device constituting the scale number management system may be distributed and realized in a plurality of devices. Further, as described above, a part or all of each device may be realized on the same device.
- the operations of the respective devices described with reference to the respective flowcharts may be stored in a computer storage device (storage medium) as the computer program of the present invention. . Then, the computer program may be read and executed by the CPU. In such a case, the present invention is constituted by the code of the computer program or a storage medium.
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
- Debugging And Monitoring (AREA)
Abstract
Description
本発明の第1の実施の形態としてのスケール数管理システム1の構成を図1に示す。図1において、スケール数管理システム1は、スケール数推定装置10と、ネットワーク機能提供装置30と、取得装置50と、制御装置70とを備える。また、スケール数推定装置10は、取得装置50および制御装置70と、それぞれ通信可能に接続されている。また、ネットワーク機能提供装置30は、取得装置50および制御装置70と、それぞれ通信可能に接続されている。また、スケール数推定装置10は、機能ルール定量化部11と、サービス量算出部12と、性能モデル生成部13と、性能モデル記憶部130と、スケール数推定部14とを有する。なお、性能モデル記憶部130は、本発明の性能モデル生成部の一部の一実施形態を構成する。また、ネットワーク機能提供装置30は、1つ以上のノード31を含む。また、ノード31内では、それぞれ、1つ以上のノードインスタンス32が稼働する。また、取得装置50は、機能ルール設定履歴記憶部501およびサービス履歴記憶部502を含む。
次に、本発明の第2の実施の形態について図面を参照して詳細に説明する。本実施の形態では、本発明におけるネットワーク機能提供装置に、サービスチェーン実行装置を適用した例について説明する。サービスチェーン実行装置は、複数のノードを連結して機能させるサービスチェーンを実行する装置である。なお、本実施の形態の説明において参照する各図面において、本発明の第1の実施の形態と同一の構成および同様に動作するステップには同一の符号を付して本実施の形態における詳細な説明を省略する。
・・・(1)
ここで、rulen1は、ノードn1の機能ルール定量値を示し、μn1は、ノードn1のサービス量を示す。このようなサービス量推定式(1)により、あるノード31の機能ルール定量値から、サービス量を推定可能となる。
ρ=λ/Sμ
・・・(2)
ここで、λは、ノード31に到着するトラフィック量(例えば、Mbps:メガバイト毎秒)を示す。また、μは、前述のサービス量推定式(1)により推定されるノード31のサービス量を示す。また、Sは、ノード31における並列処理の数であるスケール数(ノードインスタンス32の数)を示す。また、ρは、稼働率を示し、各ノード31における処理の混雑度(0~1)を表す。稼働率ρは、1に近くなるほど待ち時間が長いことを表す。稼働率ρには、規定値が設定されるものとする。
まず、取得装置50は、サービスチェーン実行装置40から、サービスチェーン43における各ノード31について機能ルール設定履歴を取得し、機能ルール設定履歴記憶部501に記憶する(ステップA1)。取得された機能ルール設定履歴は、ノードIDと、機能ルールの設定内容と、タイムスタンプとを含む。ここでは、図11に示す機能ルール設定履歴が取得されたものとする。図11では、例えばnodePの場合、タイムスタンプ「2014/03/09: 09:00:00.000」に、SSH(Secure SHell)関連で2つ、DNS(Domain Name System)関連で4つの合計6つの機能ルールが設定されている。なお、図11に示す機能ルール設定履歴は、各設定情報に、ノード31の機能のタイプを示す情報を含んでいるが、本実施の形態における機能ルール設定履歴は、少なくともノードIDと、機能ルールの設定内容と、タイムスタンプとを含んでいればよい。
次に、性能モデル生成部23は、機能ルール設定履歴記憶部501に記憶されている図11の情報から、各ノード31の設定情報毎に、機能ルールの数をカウントして機能ルール定量値とする。そして、ノードIDと、機能ルール定量値と、タイムスタンプとを対応付ける(ステップB12)。これにより、図13に示す機能ルール定量値情報が生成される。なお、図13に示す機能ルール定量値情報は、ノード31の機能のタイプを示す情報を含んでいるが、本実施の形態における機能ルール定量値情報は、少なくともノードIDと、機能ルール定量値と、タイムスタンプとを含んでいればよい。
μ=59.1/rule+20.5”・・・(1’)
で表されるサービス量推定式が求められたとする。このサービス量推定式(1’)は、機能ルール定量値(ルール数)が増えるほど、サービス量が減少することを表す(ステップB15)。
次に、チェーンインスタンス生成部24は、サービスチェーン実行装置40上に生成するサービスチェーン43に関する定義情報として、図15に示すサービスチェーン定義情報を取得する。図15では、chain1というIDで識別されるサービスチェーン43は、FWとして機能するnode1、NATとして機能するnode3、および、LBとして機能するnode7を連結して機能させるものである。また、chain2というIDで識別されるサービスチェーン43は、FWとして機能するnode2、DPI(Deep Packet Inspection)として機能するnode6を連結して機能させるものである。図15に示すように、サービスチェーン定義情報は、サービスチェーン43を構成するノードIDと、各ノード31の機能ルールの設定内容とを含む。
次に、制御装置80は、図17のチェーンインスタンス情報を読み込み(ステップD11)、サービスチェーン実行装置40上に、チェーンインスタンスを作成する。具体的には、制御装置80は、各サービスチェーン43の各ノード31について、指定されたスケール数のノードインスタンス32を生成して稼働させる。または、制御装置80は、既に稼働中のサービスチェーン43については、各ノード31のノードインスタンス32の数を、チェーンインスタンス情報に含まれるスケール数となるよう調整する(ステップD12~D14)。
10、20 スケール数推定装置
30 ネットワーク機能提供装置
40 サービスチェーン実行装置
50 取得装置
70、80 制御装置
11、21 機能ルール定量化部
12、22 サービス量算出部
13、23 性能モデル生成部
14 スケール数推定部
24 チェーンインスタンス生成部
31 ノード
32 ノードインスタンス
43 サービスチェーン
130、230 性能モデル記憶部
501 機能ルール設定履歴記憶部
502 サービス履歴記憶部
100、300、500、700 コンピュータ装置
1001、3001、5001、7001 CPU
1002、3002、5002、7002 RAM
1003、3003、5003、7003 ROM
1004、3004、5004、7004 記憶装置
1005、3005、5005、7005 ネットワークインタフェース
Claims (10)
- ネットワーク機能を提供するノードに設定された機能ルールの設定履歴に基づいて、機能ルールを定量化した機能ルール定量値を算出する機能ルール定量化手段と、
前記ノードのサービス履歴に基づいて、前記ノードにおいて稼働するノードインスタンスの単位時間あたりのサービス量を算出するサービス量算出手段と、
前記ノードについて得られた前記機能ルール定量値および前記サービス量の組に基づいて、前記ノードに対する入力トラフィック量と、前記機能ルール定量値と、前記ノードインスタンスの数(スケール数)との関係を表す性能モデルを生成する性能モデル生成手段と、
前記性能モデルを用いて、想定される前記入力トラフィック量および前記機能ルール定量値に応じた前記スケール数を推定するスケール数推定手段と、
を備えたスケール数推定装置。 - 前記性能モデル生成手段は、前記ノードについて得られた前記機能ルール定量値および前記サービス量の組に基づいて、前記機能ルール定量値から前記サービス量を算出するサービス量推定式を推定し、推定したサービス量推定式を含む前記性能モデルを生成することを特徴とする請求項1に記載のスケール数推定装置。
- 前記性能モデル生成手段は、前記サービス量推定式によって推定されるサービス量と、前記入力トラフィック量と、前記スケール数との関係を表す方程式を、前記性能モデルに含めることを特徴とする請求項2に記載のスケール数推定装置。
- 前記サービス量算出手段は、前記機能ルール定量値の有効期間毎に前記サービス量を算出することを特徴とする請求項1から請求項3のいずれか1項に記載のスケール数推定装置。
- 前記機能ルール定量化手段は、前記ノードに設定された機能ルールの数に基づいて、前記機能ルール定量値を算出することを特徴とする請求項1から請求項4のいずれか1項に記載のスケール数推定装置。
- ネットワーク機能を提供するノードからなるネットワーク機能提供装置と、
前記ネットワーク機能提供装置から、前記ノードに設定される機能ルールの設定履歴(機能ルール設定履歴)と、前記ノードのサービス履歴とを取得する取得装置と、
前記取得装置によって取得された前記機能ルール設定履歴および前記サービス履歴を用いて、前記ネットワーク機能提供装置におけるノードのスケール数を推定する請求項1から請求項5のいずれか1項に記載のスケール数推定装置と、
前記スケール数推定装置によって推定されたスケール数に基づいて、前記ネットワーク機能提供装置におけるノードのスケール数を制御する制御装置と、
を備えたスケール数管理システム。 - 前記ネットワーク機能提供装置が、複数の前記ノードを連結して機能させるサービスチェーンを実行するサービスチェーン実行装置によって構成されるとき、
前記スケール数推定装置のスケール数推定手段は、前記サービスチェーンに流入が想定される入力トラフィック量に応じて、前記サービスチェーンに含まれるノードのスケール数を推定し、
前記制御装置は、前記スケール数推定装置によって推定されたスケール数に基づいて、前記サービスチェーンにおけるノードのスケール数を制御することを特徴とする請求項6に記載のスケール数管理システム。 - ネットワーク機能を提供するノードに設定された機能ルールの設定履歴に基づいて、機能ルールを定量化した機能ルール定量値を算出し、
前記ノードのサービス履歴に基づいて、前記ノードにおいて稼働するノードインスタンスの単位時間あたりのサービス量を算出し、
前記ノードについて得られた前記機能ルール定量値および前記サービス量の組に基づいて、前記ノードに対する入力トラフィック量と、前記機能ルール定量値と、前記ノードインスタンスの数(スケール数)との関係を表す性能モデルを生成し、
前記性能モデルを用いて、想定される前記入力トラフィック量および前記機能ルール定量値に応じた前記スケール数を推定するスケール数推定方法。 - ネットワーク機能を提供するノードからなるネットワーク機能提供装置における前記ノードにおいて設定される機能ルールの設定履歴(機能ルール設定履歴)を取得し、
前記ネットワーク機能提供装置におけるノードのサービス履歴を取得し、
取得した前記機能ルール設定履歴およびサービス履歴に基づいて、請求項8に記載のスケール数推定方法を用いて、前記ネットワーク機能提供装置におけるノードのスケール数を推定し、
推定したスケール数に基づいて、前記ネットワーク機能提供装置におけるノードのスケール数を制御するスケール数管理方法。 - ネットワーク機能を提供するノードに設定された機能ルールの設定履歴に基づいて、機能ルールを定量化した機能ルール定量値を算出する機能ルール定量化ステップと、
前記ノードのサービス履歴に基づいて、前記ノードにおいて稼働するノードインスタンスの単位時間あたりのサービス量を算出するサービス量算出ステップと、
前記ノードについて得られた前記機能ルール定量値および前記サービス量の組に基づいて、前記ノードに対する入力トラフィック量と、前記機能ルール定量値と、前記ノードインスタンスの数(スケール数)との関係を表す性能モデルを生成する性能モデル生成ステップと、
前記性能モデルを用いて、想定される前記入力トラフィック量および前記機能ルール定量値に応じた前記スケール数を推定するスケール数推定ステップと、
をコンピュータ装置に実行させるコンピュータ・プログラムを記憶した記憶媒体。
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US15/509,620 US20170264500A1 (en) | 2014-09-09 | 2015-09-03 | Number-of-scales estimation apparatus, number-of-scales management system, number-of-scales estimation method, number-of-scales management method, and storage medium |
JP2016547690A JP6558374B2 (ja) | 2014-09-09 | 2015-09-03 | スケール数推定装置、スケール数管理システム、スケール数推定方法、スケール数管理方法、および、コンピュータ・プログラム |
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
JP2014-182814 | 2014-09-09 | ||
JP2014182814 | 2014-09-09 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2016038857A1 true WO2016038857A1 (ja) | 2016-03-17 |
Family
ID=55458629
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/JP2015/004477 WO2016038857A1 (ja) | 2014-09-09 | 2015-09-03 | スケール数推定装置、スケール数管理システム、スケール数推定方法、スケール数管理方法、および、記憶媒体 |
Country Status (3)
Country | Link |
---|---|
US (1) | US20170264500A1 (ja) |
JP (1) | JP6558374B2 (ja) |
WO (1) | WO2016038857A1 (ja) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017213065A1 (ja) * | 2016-06-09 | 2017-12-14 | 日本電気株式会社 | サービス管理システム、サービス管理方法、および、記録媒体 |
WO2018173481A1 (ja) * | 2017-03-24 | 2018-09-27 | 日本電気株式会社 | サービス構成設計装置、およびサービス構成設計方法 |
JPWO2020217357A1 (ja) * | 2019-04-24 | 2020-10-29 | ||
WO2024028974A1 (ja) * | 2022-08-02 | 2024-02-08 | 日本電信電話株式会社 | 性能推定モデル生成装置、性能推定装置、プログラムおよび性能推定モデル生成方法 |
Families Citing this family (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
ES2824681T3 (es) * | 2015-09-30 | 2021-05-13 | Huawei Tech Co Ltd | Método, dispositivo y sistema de ampliación |
EP3745761A1 (en) * | 2019-05-28 | 2020-12-02 | Samsung Electronics Co., Ltd. | Virtualization of ran functions based on load of the base stations |
US11219032B2 (en) * | 2019-05-28 | 2022-01-04 | Samsung Electronics Co., Ltd. | Method and apparatus for performing function of radio access network |
CN116584079A (zh) * | 2020-12-04 | 2023-08-11 | 三星电子株式会社 | 执行无线接入网络功能的方法和装置 |
Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012074056A (ja) * | 2011-11-10 | 2012-04-12 | Telecom Italia Spa | 通信サービス及びネットワーク管理のうちの一方または双方のプラットフォームにおけるリソース管理ための方法、対応するプラットフォーム及びコンピュータ・プログラム製品 |
JP2013179429A (ja) * | 2012-02-28 | 2013-09-09 | Nippon Telegr & Teleph Corp <Ntt> | 振分先を切り替える並列パケット処理方法および装置 |
-
2015
- 2015-09-03 US US15/509,620 patent/US20170264500A1/en not_active Abandoned
- 2015-09-03 WO PCT/JP2015/004477 patent/WO2016038857A1/ja active Application Filing
- 2015-09-03 JP JP2016547690A patent/JP6558374B2/ja active Active
Patent Citations (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2012074056A (ja) * | 2011-11-10 | 2012-04-12 | Telecom Italia Spa | 通信サービス及びネットワーク管理のうちの一方または双方のプラットフォームにおけるリソース管理ための方法、対応するプラットフォーム及びコンピュータ・プログラム製品 |
JP2013179429A (ja) * | 2012-02-28 | 2013-09-09 | Nippon Telegr & Teleph Corp <Ntt> | 振分先を切り替える並列パケット処理方法および装置 |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2017213065A1 (ja) * | 2016-06-09 | 2017-12-14 | 日本電気株式会社 | サービス管理システム、サービス管理方法、および、記録媒体 |
WO2018173481A1 (ja) * | 2017-03-24 | 2018-09-27 | 日本電気株式会社 | サービス構成設計装置、およびサービス構成設計方法 |
JPWO2018173481A1 (ja) * | 2017-03-24 | 2020-01-30 | 日本電気株式会社 | サービス構成設計装置、およびサービス構成設計方法 |
JPWO2020217357A1 (ja) * | 2019-04-24 | 2020-10-29 | ||
WO2020217357A1 (ja) * | 2019-04-24 | 2020-10-29 | 三菱電機株式会社 | 情報処理システム |
CN113678109A (zh) * | 2019-04-24 | 2021-11-19 | 三菱电机株式会社 | 信息处理系统 |
JP7184170B2 (ja) | 2019-04-24 | 2022-12-06 | 三菱電機株式会社 | 情報処理システム |
WO2024028974A1 (ja) * | 2022-08-02 | 2024-02-08 | 日本電信電話株式会社 | 性能推定モデル生成装置、性能推定装置、プログラムおよび性能推定モデル生成方法 |
Also Published As
Publication number | Publication date |
---|---|
JP6558374B2 (ja) | 2019-08-14 |
US20170264500A1 (en) | 2017-09-14 |
JPWO2016038857A1 (ja) | 2017-06-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
JP6558374B2 (ja) | スケール数推定装置、スケール数管理システム、スケール数推定方法、スケール数管理方法、および、コンピュータ・プログラム | |
US9967188B2 (en) | Network traffic flow management using machine learning | |
US10027739B1 (en) | Performance-based content delivery | |
JP6493400B2 (ja) | サービスチェーン管理装置、サービスチェーン管理システム、サービスチェーン管理方法、及び、プログラム | |
CN108429701B (zh) | 网络加速系统 | |
Benedetti et al. | Reinforcement learning applicability for resource-based auto-scaling in serverless edge applications | |
CN109040186A (zh) | 一种基于nbiot网络的mqtt数据处理方法和装置 | |
JP6475966B2 (ja) | ネットワーク設計装置及びプログラム | |
JP6660283B2 (ja) | トラヒック需要予測装置、トラヒック需要予測方法、及びプログラム | |
JP5917678B1 (ja) | 情報処理装置、方法およびプログラム | |
Zinner et al. | A discrete-time model for optimizing the processing time of virtualized network functions | |
CN109194545A (zh) | 一种网络试验平台流量生成系统、方法、装置及电子设备 | |
CN112436951B (zh) | 一种预知流量路径的方法和装置 | |
US10447567B2 (en) | Control apparatus and processing method for control apparatus | |
CN108718259B (zh) | 一种报文处理方法及多核处理器 | |
JP6554011B2 (ja) | タイムアウト時間設定装置とタイムアウト時間設定方法 | |
US20180150375A1 (en) | Service management system, service management method, and recording medium | |
JPWO2018173481A1 (ja) | サービス構成設計装置、およびサービス構成設計方法 | |
Munir et al. | Planning data transfers in grids: a multi‐service queueing approach | |
JP6070717B2 (ja) | 分散データ処理システム、及び、分散データ処理方法 | |
JP7095624B2 (ja) | 識別装置及び識別プログラム | |
US20230308392A1 (en) | Linked Packet Tracing for Software Load Balancers | |
JP4401214B2 (ja) | 負荷分散装置及びプログラム | |
WO2019239759A1 (ja) | データ送信方法、データ送信装置及びプログラム | |
KR101563457B1 (ko) | 서비스 가용성을 보장하기 위한 이웃 노드 선출 시스템 및 방법 |
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: 15839570 Country of ref document: EP Kind code of ref document: A1 |
|
ENP | Entry into the national phase |
Ref document number: 2016547690 Country of ref document: JP Kind code of ref document: A |
|
WWE | Wipo information: entry into national phase |
Ref document number: 15509620 Country of ref document: US |
|
NENP | Non-entry into the national phase |
Ref country code: DE |
|
122 | Ep: pct application non-entry in european phase |
Ref document number: 15839570 Country of ref document: EP Kind code of ref document: A1 |