Data processing device, method, and program

Abstract

A data processing device according to an embodiment comprises: an input unit for receiving an input of request information which pertains to the provision of a service provided by using a communication network and which is requested by a user who uses the service; a first identification unit for identifying a category to which the service indicated by the request information belongs and which includes information about a resource necessary for providing the service indicated by the information; a second identification unit for identifying, on the basis of the information about the necessary resource, the allocation of the resource to the user to which the service is provided; a first calculation unit for calculating, on the basis of the allocation of the resource, an index of the quality for the service corresponding to the category; a second calculation unit for calculating, on the basis of the index calculated by the first calculation unit, a user's quality of experience value for the provided service; a third calculation unit for calculating, on the basis of the allocation of the resource, the cost pertaining to the provision of the service; and an optimization unit for optimizing the allocation of the resource to the user so that the difference between the quality of experience value and the cost is optimized.

Description

Data processing device, method and program

[0001] FIELD Embodiments of the present invention relate to a data processing device, a method, and a program.

[0002] In an Intent-Based Network (IBN), users declaratively specify service requirements (hereinafter sometimes referred to as "requirements") as intents, i.e., requests for service provision, in order to autonomously control resources, including communication networks and computational resources deployed on the networks. Based on the specified intents, the system operates autonomously to fulfill the intents.

[0003] Furthermore, because an intent does not specify a specific procedure for fulfilling the intent, the system operates to fulfill the declared intent even in the face of dynamic network changes.

[0004] In this way, in order for the system to operate based on an intent, interactive interaction between the system and the user is necessary for the purpose of completing information necessary to define the intent.

[0005] Furthermore, due to the limited availability and performance of physical and virtual resources, conflicts of intent may occur between multiple users and systems. To resolve such conflicts, interactive intent negotiation between the system and the user is being considered.

[0006] Non-Patent Document 1 discloses a method for allocating resources to satisfy an intent from among resources available at the time of intent processing. In this method, if resources to satisfy the intent cannot be secured, a second-best intent is presented to confirm whether it can be accepted.

[0007] In addition, in Non-Patent Document 2, priorities are set for intents, and the linear sum of the priorities of each intent is maximized while satisfying resource constraints. As described above, in the conventional technology, conflicts between intents are resolved based on factors outside the intents.

[0008] That is, the method disclosed in Non-Patent Document 1 resolves intent conflicts on a first-come, first-served basis, while the method disclosed in Non-Patent Document 2 resolves intent conflicts on a priority basis.

[0009] Y. Sharma, D. Bhamare, A. Kassler and J. Taheri, "Intent Negotiation Framework for Intent-Driven Service Management," in IEEE Communications Magazine, vol. 61, no. 6, pp. 73-79, June 2023, doi: 10.1109 / MCOM.001.2200504.Q. Zhang, J. Chen, D. Gao and X. Wang, "Intent-based Service Policy Conflict Management Algorithm," 2022 IEEE / CIC International Conference on Communications in China (ICCC), Sanshui, Foshan, China, 2022, pp. 19-24, doi: 10.1109 / ICCC55456.2022.9880783.

[0010] As described above, in conventional technologies, conflict resolution between intents is performed based on factors outside the intent, so the results of intent negotiation that presents alternative intents to the user are not included in the intent, and the intent cannot be continuously fulfilled in the face of dynamic network changes.

[0011] Specifically, if an alternative intent is accepted due to a lack of resources during negotiation, the system will not be able to allocate surplus resources even if the availability of resources is restored, because the system only retains information about the alternative intent.

[0012] The present invention has been made in light of the above circumstances, and an object of the present invention is to provide a data processing device, method, and program that are capable of appropriately adjusting resources.

[0013] A data processing device according to one aspect of the present invention comprises an input unit that accepts input of request information regarding the provision of a service from a user who uses a service provided using a communication network; a first identification unit that identifies a category to which the service indicated by the request information belongs, including information on resources required to provide the service indicated by the request information; a second identification unit that identifies a resource allocation to the user to whom the service is to be provided based on the information on the required resources; a first calculation unit that calculates a quality index for the service corresponding to the category based on the resource allocation; a second calculation unit that calculates a value of the quality of experience of the service provided by the user based on the index calculated by the first calculation unit; a third calculation unit that calculates a cost associated with the provision of the service based on the resource allocation; and an optimization unit that optimizes the resource allocation to the user so as to maximize the difference between the quality of experience value and the cost.

[0014] A data processing method according to one aspect of the present invention is a method performed by a data processing device, and includes: an input unit of the data processing device accepting input of information regarding the provision of a service from a user who uses a service provided using a communication network; a first identification unit of the input unit of the data processing device identifying a category to which the service indicated by the information belongs, the category including information on resources required to provide the service indicated by the information on the requested resources; a second identification unit of the data processing device identifying an allocation of resources to the user to whom the service is to be provided based on the information on the required resources; a first calculation unit of the data processing device calculating a quality index for the service corresponding to the category based on the resource allocation; a second calculation unit of the data processing device calculating a value of the quality of experience of the service provided by the user based on the index calculated by the first calculation unit; a third calculation unit of the data processing device calculating a cost related to the provision of the service based on the resource allocation; and an optimization unit of the data processing device optimizing the allocation of resources to the user so as to maximize the difference between the quality of experience value and the cost.

[0015] According to the present invention, resources can be appropriately adjusted.

[0016] FIG. 1 is a diagram illustrating an example of an intent. FIG. 2 is a diagram illustrating an application example of a data processing device according to an embodiment of the present invention. FIG. 3 is a diagram illustrating each process performed by the data processing device according to this embodiment. FIG. 4 is a flowchart illustrating an example of a processing operation performed by the data processing device according to this embodiment. FIG. 5 is a diagram illustrating an example of QoE characteristics with respect to delay time. FIG. 6 is a diagram illustrating an example of QoE characteristics with respect to delay time. FIG. 7 is a diagram illustrating an example of a network topology to which this embodiment is applied. FIG. 8 is a diagram illustrating an example of the number of intents and the ratio of accepted intents. FIG. 9 is a block diagram illustrating an example of the hardware configuration of a data processing device according to an embodiment of the present invention.

[0017] An embodiment of the present invention will be described below. Fig. 1 is a diagram illustrating an example of an intent. In existing intent negotiation, a single condition desired by a user, such as a condition in the method disclosed in the above-mentioned Non-Patent Document 1, or a minimum threshold that must be satisfied, such as a condition in the method disclosed in the above-mentioned Non-Patent Document 2, is retained as an intent obtained as a result of intent negotiation. In the example shown in Fig. 1(a), the intent includes a service category, a maximum number of simultaneous connections, a delay time, and a cost.

[0018] Therefore, as shown in FIG. 1A, even if a resource becomes available as a result of intent negotiation, only the intent after negotiation can be referenced.

[0019] In contrast, in this embodiment, instead of a single condition related to requirements and cost, as shown in FIG. 1B, a function of Quality of Experience (QoE), which is a value of the user's perceived quality of the service provided and calculated from resource allocation, is considered as a characteristic of requirements and cost, and is accepted as an intent that includes the negotiation process. The QoE characteristics shown in FIG. 1B are fitted from the intent specified and approved by the user. In this embodiment, when the network dynamically changes, such as when resource availability changes, resource allocation is determined according to the QoE characteristics. This QoE characteristic differs depending on the type of user or service.

[0020] Fig. 2 is a diagram showing an application example of a data processing device according to an embodiment of the present invention. Fig. 3 is a diagram explaining each process performed by the data processing device according to this embodiment. Fig. 4 is a flowchart showing an example of a processing operation performed by the data processing device according to this embodiment. In the example shown in Fig. 2, the data processing device 100 according to this embodiment includes a storage unit 11 having a storage device such as a nonvolatile memory, an intent translation unit 12, an intent completion unit 13, a service category identification unit 14, an information extraction unit 15, a resource allocation unit 16, a QoS calculation unit 17, a QoE calculation unit 18, a cost calculation unit 19, and a resource allocation optimization unit 20.

[0021] In this embodiment, the QoE and cost of resource allocation are modeled. The data processing device 100 defines a trade-off between the QoE and cost of the service and the user, and maximizes the QoE-cost, which is the difference between the utility of each intent, i.e., the QoE cost, in resource adjustment during intent conflict.

[0022] In the IBN, users are defined as owners of services operated on various networks, and service providers are defined as providers of resources, including networks and computational resources deployed on the networks. When intent translation is necessary (Yes in S11), the intent translation unit 12 performs intent translation (S12). When intent information is insufficient (Yes in S13), the intent completion unit 13 performs intent completion (S14). Through these translations and completions, the service category identification unit 14 refers to the knowledge base stored in the storage unit 11 (symbol a in FIG. 3 ) to identify a service category (S15).

[0023] This service category includes the resources required to provide the service, i.e., the type and amount of resources required for each function in the SFC (Service Function Chain). Here, resources refer to network bandwidth and computational resources such as a CPU (Central Processing Unit), memory, and storage. When the intent is that of a service provider, the utility of the intent is calculated based only on the identified resource allocation.

[0024] As shown in (1) below, the service category includes parameters of a Quality of Service (QoS) function and parameters of a QoE function. QoS is an index of the quality of a service.

[0025]

[0026] After the service category is identified from the intent, the information extraction unit 15 extracts the necessary resources for the SFC, the parameters of the QoS function, and the parameters of the QoE function (symbol b in FIG. 3).

[0027] Then, in accordance with the extracted required resources of the SFC, the resource allocation unit 16 specifies a resource allocation for the user in resource k of node n on the network. Here, the resource allocation is specified arbitrarily within a range that satisfies the resource requirements of the SFC and the resource capacity on the network.

[0028] The QoS calculation unit 17 calculates one or more QoS values ​​corresponding to the service category from the identified resource allocation in accordance with the QoE function, as shown in (2) below. That is, the QoS value is calculated using the resource allocation as an argument of the QoE function.

[0029]

[0030] Then, the QoE calculation unit 18 calculates one QoE for each user from the calculated QoS according to the QoE function, as shown in (3) below (S16). That is, the QoE value is calculated using the calculated QoS as an argument of the QoE function, and the parameters related to the user are updated.

[0031]

[0032] Furthermore, the cost calculation unit 19 calculates the cost for each user from the unit price c according to the amount of resource usage related to the resource allocation, as shown in (4) below.

[0033]

[0034] The QoE and cost values ​​correspond to each other so that they can be compared. Then, the resource allocation optimization unit 20 optimizes the resource allocation so as to maximize the difference between the QoE and the cost, as shown in (5) below (S17).

[0035]

[0036] This optimized resource allocation is presented to the user, and if approval is not obtained from the user (No in S18), an alternative intent is accepted (S19) and the process returns to S11. On the other hand, if approval is obtained (Yes in S18), the optimized resource allocation is implemented (S21) and user-related parameters are updated (S22).

[0037] In this embodiment, by modeling QoE and cost, conflicts between intents are resolved based on the trade-off between them. This allows resources between intents to be adjusted by retaining and referencing these characteristics even when a new intent arrives or the network changes dynamically.

[0038] Examples of the above QoS function and QoE function will be described. First, as an example of QoS, an example of the delay time shown in (6) below will be shown. In this example shown in (6), the QoS function is a function characterized by the parameters of the QoS function, with the resource allocation to the user as an argument.

[0039]

[0040] The delay time in a Web conference service can be divided into a communication delay in the network and a delay in the computing node, for example, a processing delay. This network delay is calculated by multiplying the communication data size d ｎｅｔｗｏｒｋ The network bandwidth l when the link on the network is e ｅ This is the value divided by .

[0041] The proportion of the network delay in the overall service delay differs for each SFC function, so the network delay is corrected using the parameter shown in (7) below.

[0042]

[0043] The delay of the computing node is calculated by multiplying the processing data size dcomputation by the processing capacity a of the computing resource. ｎ，ｋSince the effect of the delay of the computing node on the processing time differs for each resource, the delay of the computing node is corrected by the parameter shown in (8) below.

[0044]

[0045] That is, in this delay time example, resources are allocated to the computational resources and network bandwidth of each node. The parameters defined for each service represent the impact of each resource on the delay time. In addition to the above example, other types of QoS with other function forms, parameters, and types are possible.

[0046] Next, as an example of QoE, a sigmoid function in delay time is shown in the following (9). In the example shown in (9), the QoE function is a function characterized by the parameters α and β of the QoE function, with QoS as an argument.

[0047]

[0048] 5 and 6 are diagrams showing examples of QoE characteristics with respect to delay time. Fig. 5 shows the QoE characteristics for a service with a relatively low priority and insensitive to QoS, while Fig. 6 shows the QoE characteristics for a service with a relatively high priority and sensitive to QoS. The characteristics shown in Fig. 5 and 6 correspond to the same QoE but with different QoS, i.e., different allocated resources.

[0049] As shown in Figures 5 and 6, the QoE characteristics relative to delay time as QoS are nonlinear, and it is known that in the range of relatively small delay and the range of relatively large delay, the amount of change in QoE per unit delay time decreases compared to other ranges.

[0050] That is, the user cannot feel the change in QoE when the delay time is sufficiently small, and similarly, the user cannot feel the effect of the change in QoE when the delay time is sufficiently large.

[0051] Although a standardized QoE can be used, in this embodiment, a scaled QoE is used so that it can be compared with the cost. In addition to the above example, other function forms, parameters, and types of QoE can be considered.

[0052] FIG. 7 is a diagram showing an example of a network topology to which this embodiment is applied. FIG. 7 shows a fat tree type network topology. Here, an example will be described in which resources are allocated to multiple intents that specify two types of service categories in this network topology. Here, a network topology consisting of four core servers, 16 edge servers, and 32 hosts is taken as an example.

[0053] The resource capacity of the upper server is larger than that of the lower server. The service specified by the intent is provided through the host, and if it cannot be deployed to the host due to resource constraints, it is deployed to the upper server. For simplicity, we will assume that there is one function of the SFC and one type of resource, and we will consider only computational resources without considering network resources.

[0054] In addition, the correspondence between resource allocation and QoE is approximated by a piecewise linear function of two pieces. The baseline of the existing technology accepts intents one by one, and repeatedly allocates resources to the intents so as to maximize the requested QoE. If resources cannot be allocated to an intent, the intent is rejected.

[0055] 8 is a diagram illustrating an example of the number of intents and the ratio of accepted intents. FIG. 8 illustrates an example of the ratio of intents accepted by each technique when the number of intents is changed. Compared to the acceptance ratio in the baseline (symbol a in FIG. 8 ), the acceptance ratio in this embodiment (symbol b in FIG. 8 ) consistently improves regardless of the number of intents.

[0056] 9 is a block diagram showing an example of the hardware configuration of a data processing device according to an embodiment of the present invention. In the example shown in FIG. 9, the data processing device 100 according to the embodiment is configured, for example, as a server computer or a personal computer, and has a hardware processor 111A such as a CPU. A program memory 111B, a data memory 112, an input / output interface 113, and a communication interface 114 are connected to this hardware processor 111A via a bus 115.

[0057] The communication interface 114 includes, for example, one or more wireless communication interface units, and enables transmission and reception of information to and from a communication network. As the wireless interface, for example, an interface that adopts a low-power wireless data communication standard such as a wireless LAN (Local Area Network) is used.

[0058] An input device 300 and an output device 400 attached to the data processing device 100 and used by a user or the like are connected to the input / output interface 113. The input / output interface 113 receives operation data input by a user or the like through the input device 300 such as a keyboard, a touch panel, a touchpad, or a mouse, and outputs output data to an output device 400 including a display device using a liquid crystal or an organic electroluminescence (EL) display, for display. The input device 300 and the output device 400 may be devices built into the data processing device 100, or may be input devices and output devices of other information terminals that can communicate with the data processing device 100 via the network NW.

[0059] The program memory 111B is a non-transitory tangible storage medium that is a combination of a non-volatile memory that can be written to and read from at any time, such as a hard disk drive (HDD) or a solid state drive (SSD), and a non-volatile memory such as a read only memory (ROM), and stores programs necessary to execute various control processes, etc., according to one embodiment.

[0060] The data memory 112 is a tangible storage medium that is, for example, a combination of the above-mentioned nonvolatile memory and a volatile memory such as RAM (Random Access Memory), and is used to store various data acquired and created in the course of various processes performed by the data processing device 100.

[0061] A data processing device 100 according to an embodiment of the present invention may be configured as a data processing device having software-based processing function units, i.e., the units shown in Fig. 2. Storage areas used as work memory or the like by the units of the data processing device 100 may be configured using the data memory 112 shown in Fig. 9. However, these configured storage areas are not essential components within the data processing device 100, and may be areas provided in, for example, an external storage medium such as a USB (Universal Serial Bus) memory, or a storage device such as a database server located in the cloud.

[0062] The processing function unit can be realized by having the hardware processor 111A read and execute a program stored in the program memory 111B, but the processing function unit may also be realized in various other forms, including an integrated circuit such as an application specific integrated circuit (ASIC) or a field-programmable gate array (FPGA).

[0063] The methods described in each embodiment can be stored as a program (software means) that can be executed by a computer on a recording medium such as a magnetic disk (floppy disk, hard disk, etc.), optical disk (CD-ROM, DVD, MO, etc.), or semiconductor memory (ROM, RAM, flash memory, etc.), and can also be distributed by transmitting it via a communication medium. The program stored on the medium also includes a configuration program that configures the software means (including not only execution programs but also tables and data structures) that the computer executes. The computer that realizes this device reads the program stored on the recording medium and, in some cases, configures the software means using the configuration program, and executes the above-mentioned processing by controlling the operation of this software means. The term "recording medium" as used herein is not limited to a storage medium for distribution, but also includes storage media such as a magnetic disk or semiconductor memory installed inside the computer or in a device connected via a network.

[0064] The present invention is not limited to the above-described embodiments, and various modifications can be made in the implementation stage without departing from the spirit of the invention. Furthermore, the embodiments may be implemented in appropriate combinations, in which case the combined effects can be obtained. Furthermore, the above-described embodiments include various inventions, and various inventions can be extracted by combining selected elements from the disclosed elements. For example, if the problem can be solved and the desired effect can be obtained even if some elements are deleted from all elements shown in the embodiments, the configuration from which these elements are deleted can be extracted as an invention.

[0065] REFERENCE SIGNS LIST 100: Data processing device 11: Storage unit 12: Intent translation unit 13: Intent completion unit 14: Service category identification unit 15: Information extraction unit 16: Resource allocation unit 17: QoS calculation unit 18: QoE calculation unit 19: Cost calculation unit 20: Resource allocation optimization unit

Claims

1. A data processing device comprising: an input unit that accepts input of request information regarding the provision of a service from a user who uses a service provided using a communication network; a first identification unit that identifies a category to which the service indicated by the request information belongs, including information on resources necessary to provide the service indicated by the information; a second identification unit that identifies an allocation of resources to the user to whom the service is to be provided based on the information on the necessary resources; a first calculation unit that calculates a quality index for the service corresponding to the category based on the resource allocation; a second calculation unit that calculates a value of the quality of experience of the service provided by the user based on the index calculated by the first calculation unit; a third calculation unit that calculates a cost related to the provision of the service based on the resource allocation; and an optimization unit that optimizes the allocation of resources to the user so as to maximize the difference between the quality of experience value and the cost.

2. The data processing device of claim 1, wherein the category includes a function for calculating a quality index for the service and a function for calculating a quality-of-experience value for the provided service, the first calculation unit calculates a quality index for the service corresponding to the category by using the resource allocation as an argument of the function for calculating a quality index for the service, and the second calculation unit calculates a quality-of-experience value for the provided service by using the index calculated by the first calculation unit as an argument of the function for calculating a quality-of-experience value for the provided service.

3. The data processing device according to claim 1, wherein the quality index for the service corresponding to the category is the sum of a communication delay time in the communication network and a processing delay time by a node used to provide the service.

4. The data processing device according to claim 3, wherein the second calculation unit calculates the quality of experience value of the provided service based on a nonlinear function having as arguments the sum of the communication delay time in the communication network and the processing delay time by the node used to provide the service.

5. A method performed by a data processing device, comprising: an input unit of the data processing device accepting input of information on requests for service provision from a user who uses a service provided via a communications network; a first identification unit of the input unit of the data processing device identifying a category to which the service indicated by the information belongs, the category including information on resources required to provide the service indicated in the information; a second identification unit of the data processing device identifying an allocation of resources to the user to whom the service is to be provided based on the information on the required resources; a first calculation unit of the data processing device calculating a quality index for the service corresponding to the category based on the resource allocation; a second calculation unit of the data processing device calculating a value of the quality of experience of the service provided by the user based on the index calculated by the first calculation unit; a third calculation unit of the data processing device calculating a cost related to the provision of the service based on the resource allocation; and an optimization unit of the data processing device optimizing the allocation of resources to the user so as to maximize the difference between the quality of experience value and the cost.

6. A data processing program that causes a processor to function as each part of the data processing device according to any one of claims 1 to 4.

Images