Routing information provision device and program

Abstract

In order to facilitate provision of routing information for eliminating congestion occurring in a network while suppressing a deterioration in communication quality, this routing information provision device comprises: a creation unit that creates input information on the basis of information pertaining to congestion in a network, information pertaining to the network, and QoS requirements of the network; and a provision unit that, on the basis of the input information, provides routing information for avoiding congestion in the network.

Description

Routing information providing device and program

[0001] The present invention relates to a routing information providing device and a program.

[0002] Fluctuations in network usage can cause congestion in the network, resulting in a decline in communication quality. There are also technologies available to resolve congestion when it occurs.

[0003] For example, a congestion avoidance algorithm is known that performs control such as reducing the amount of packets sent depending on the QoS (Quality of Service) status of the network (see, for example, Non-Patent Document 1).

[0004] Also, a routing algorithm is known that predicts the required quality for each service or user, and controls communication paths in advance based on the prediction results to avoid congestion (see, for example, Non-Patent Document 2).

[0005] M. Hock, R. Bless and M. Zitterbart, "Experimental evaluation of BBR congestion control," IEEE 25th International Conference on Network Protocols (ICNP), pp. 1-10, 2017. Kobayashi and Harada, "Network control technology based on quality of service requirements," IEICE Technical Report, vol. 120, no. 314, CQ2020-64, pp. 22-26, 2021.

[0006] The congestion avoidance algorithm as disclosed in Non-Patent Document 1 avoids congestion by limiting the amount of packets, which has the problem of lowering QoS such as throughput.

[0007] On the other hand, when the network becomes complex, the routing algorithm shown in Non-Patent Document 2 requires solving a combinatorial problem to calculate a solution, which can lead to a computational explosion. For this reason, various approximation algorithms have been proposed, but they have the problem that they are not necessarily suited to the current network usage situation of users or the target network.

[0008] One method is for an operator with specialized knowledge to change the routing depending on the situation, but in the case of large-scale networks, it is difficult to find the optimal control, and skilled experts are required.

[0009] As described above, when congestion occurs in a network, it is difficult to resolve the congestion with conventional techniques.

[0010] The embodiments of the present invention have been made in view of the above-mentioned problems, and make it possible to easily provide routing information for resolving congestion when congestion occurs in a network.

[0011] In order to solve the above problem, a routing information providing device according to an embodiment of the present invention has a creation unit that creates input information based on congestion information, network information, and QoS requirements, and an acquisition unit that inputs the input information into the generation unit that generates routing information for avoiding the congestion based on the input information and acquires the routing information.

[0012] According to an embodiment of the present invention, when congestion occurs in a network, routing information for resolving the congestion can be easily provided.

[0013] It is a diagram showing an example of the configuration of a routing information providing system according to the present embodiment.It is a flowchart showing an example of routing information providing processing according to the present embodiment.It is a diagram showing an example of the hardware configuration of a computer.

[0014] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. The embodiment described below is merely an example, and the embodiment to which the present invention is applied is not limited to the following embodiment.

[0015] <System Configuration> Fig. 1 is a diagram showing an example of the configuration of a routing information providing system according to this embodiment. The routing information providing system 1 is a system that provides, when congestion occurs in a network, routing information indicating a communication path for resolving congestion. Preferably, the routing information providing system 1 provides routing information for resolving congestion while suppressing a decrease in QoS.

[0016] In the example of FIG. 1, the routing information providing system 1 includes a routing information providing device 100 , a monitoring system 10 , an administrator terminal 20 , and an LLM 30 that can communicate with the routing information providing device 100 .

[0017] The monitoring system 10 is a system that monitors information about congestion, including, for example, information about the location of congestion when congestion occurs, time-series data of the network QoS value, and / or time-series data of the network user usage status. The information about the location of congestion includes, for example, information such as the node name or link name of the location where congestion occurred. The network QoS value includes, for example, information such as delay, packet loss rate, or throughput. The user usage status includes, for example, fluctuations in the number of users or fluctuations in packet communication volume. Note that, in this embodiment, an existing system can be applied to the monitoring system 10.

[0018] The administrator terminal 20 is an information processing device such as a personal computer (PC), tablet terminal, smartphone, or dedicated terminal used by an administrator who manages the network. The administrator uses the administrator terminal 20 or the like to set network information such as network configuration information, network bandwidth setting information, and routing setting candidate information, as well as QoS requirements, in advance in the routing information providing device 100. The administrator terminal 20 also displays routing information provided by the routing information providing device 100. Note that the administrator terminal 20 that displays information related to the network and the administrator terminal 20 that displays routing information may be the same terminal or different terminals.

[0019] The LLM (generation unit) 30 is a large-scale language model that generates routing information for avoiding network congestion based on input information (prompts) created by the routing information providing device 100. The large-scale language model is a language model constructed using a large amount of text data and deep learning technology.

[0020] In this embodiment, information about network congestion, information about the network, QoS requirements, etc. are input as "prompts" in text format to the LLM 30, which then outputs routing information (routing solution) for avoiding congestion. This automatically obtains routing information corresponding to the corresponding network and the current user usage status. The LLM 30 is an example of a generation unit that generates routing information for avoiding congestion based on input information.

[0021] The routing information providing device 100 is an information processing device having a computer configuration, or a system including multiple computers. The routing information providing device 100 realizes each functional configuration shown in Fig. 1 by, for example, executing a predetermined program on a computer provided in the routing information providing device 100. In the example of Fig. 1, the routing information providing device 100 has each functional configuration such as an input unit 101, an information aggregating unit 102, a creating unit 103, an acquiring unit 104, a calculating unit 105, a determining unit 106, a providing unit 107, and a storage unit 108. Note that at least a part of each of the above functional configurations may be realized by hardware.

[0022] The input unit 101 executes an input process for accepting input (setting) of the above-described network information and QoS requirements from the administrator terminal 20. For example, the input unit 101 provides the administrator terminal 20 with a UI (User Interface) or an API (Application Programming Interface) for inputting the network information and QoS information.

[0023] When congestion occurs, the information aggregating unit 102 outputs a "congestion status summary text" that expresses the status of the congestion-related information obtained from the monitoring system 10 in text data format. Note that text is defined as a sequence of multiple sentences. For example, the information aggregating unit 102 converts the congestion-related information obtained from the monitoring system 10 into a congestion status summary text based on a preset common template.

[0024] The creation unit 103 executes a creation process to create input information based on congestion information, network information, and QoS requirements. For example, the creation unit 103 creates input information that is a prompt that instructs the LLM (Large Scale Language Model) 30 to generate routing information for avoiding congestion.

[0025] For example, the creation unit 103 creates a prompt to be input to the LLM 30 using the congestion status summary text output by the information collection unit 102, network information preset by the administrator, and QoS requirements as input.

[0026] The acquisition unit 104 inputs the input information created by the creation unit 103 to the LLM (generation unit) 30, and executes an acquisition process to acquire routing information for avoiding congestion.

[0027] The calculation unit 105 executes a calculation process to calculate the current network QoS value based on information about the network, the user usage status of the network, and the routing information acquired by the acquisition unit 104. Note that the QoS value may be calculated using a known QoS calculation system, or may be calculated by the routing information providing device 100 using a known calculation formula.

[0028] The determination unit 106 executes a determination process to determine whether the current network QoS value calculated by the calculation unit 105 satisfies the QoS requirements. If the current network QoS value does not satisfy the QoS requirements, the creation unit 103 creates different input information that instructs the LLM 30 to generate different routing information, and the acquisition unit 104 inputs the different input information to the LLM 30 to acquire the different routing information.

[0029] If the current network QoS value satisfies the QoS requirements, the providing unit 107 executes a providing process to provide the routing information acquired by the acquiring unit 104 to a predetermined destination such as the administrator terminal 20.

[0030] The storage unit 108 is a storage unit that can be accessed from each of the above functional configurations. The storage unit 108 is realized, for example, by a storage device of a computer included in the routing information providing device 100 and a program executed by the computer.

[0031] 1 is an example. For example, the calculation unit 105 or the storage unit 108 included in the routing information providing device 100 in FIG. 1 may be provided outside the routing information providing device 100. Furthermore, the LLM 30 may be provided inside the routing information providing device 100.

[0032] <Processing Flow> Next, the processing flow of the routing information providing method according to this embodiment will be described.

[0033] 2 is a flowchart showing an example of a routing information providing process according to this embodiment. This process shows an example of a process executed by the routing information providing device 100 described in FIG. 1 when congestion occurs in the network. It is assumed that, at the start of the process shown in FIG. 2, the administrator has already set the network information and QoS requirements described above in the routing information providing device 100 using the administrator terminal 20 or the like.

[0034] In step S201, the information aggregation unit 102 collects data indicating the situation at the time of congestion, which is obtained through the monitoring system 10, and stores the collected data in the storage unit 108 or the like. Note that the data indicating the situation at the time of congestion includes, for example, information on the location of congestion, time-series data on QoS, and time-series data on user usage status. Note that the time-series data may have a time-series length of any length before and after the time of congestion occurrence.

[0035] In step S202, the information aggregating unit 102 converts data representing the situation at the time of congestion, stored in the storage unit 108, etc., into congestion information aggregation text, which is data explaining the situation in text format. Note that a common text template is provided to determine how the data representing the situation at the time of congestion is converted into text.

[0036] In step S203, the creation unit 103 creates input information based on the congestion status summary text converted by the information aggregation unit 102, the network information preset by the administrator, and the QoS conditions. For example, the creation unit 103 sequentially writes the congestion status summary text as supplemental information to the LLM 30, and based on the written supplemental information, creates input information that is a prompt to the LLM 30, which writes an instruction to find routing information (routing solution) that conferencing the congestion.

[0037] In step S204, the acquisition unit 104 acquires routing information for avoiding congestion based on the input information created by the creation unit 103. For example, when the acquisition unit 104 inputs the input information (prompt) created by the creation unit 103 to the LLM 30, the LLM 30 responds with routing information (routing solution) for avoiding congestion.

[0038] By performing the processes of steps S201 to S204, the routing information providing device 100 can easily obtain routing information for resolving congestion when congestion occurs in the network.

[0039] However, in the case of a complex network, the LLM 30 may often give an incorrect answer and output incorrect information (false information). In such cases, the correct result can often be reached by repeating the question and answer process while pointing out the error. Therefore, the routing information providing device 100 according to this embodiment further executes the process from step S205 onwards.

[0040] In step S205, the calculation unit 105 calculates a current QoS value based on the network information, the user's network usage status, and the acquired routing information. For example, the calculation unit 105 calculates the current QoS value by performing routing based on the acquired routing information based on the network information set by the administrator and the user's usage status collected by the monitoring system 10. Here, the QoS value includes, for example, values ​​such as the current delay, packet loss rate, or throughput. Note that a general calculation formula can be applied to calculate the QoS value.

[0041] In step S206, the determination unit 106 determines whether the current QoS value calculated by the calculation unit 105 satisfies the QoS requirements. If the QoS requirements are not satisfied, the determination unit 106 proceeds to step S207. On the other hand, if the QoS requirements are satisfied, the determination unit 106 proceeds to step S208.

[0042] In step S207, the creation unit 103 creates input information (prompt) stating that the QoS requirements are not met, and returns the process to step S204. For example, the creation unit 103 adds the calculated QoS value as additional text to the input information created in step S203 and creates another input information stating that the QoS requirements are not met. The acquisition unit 104 inputs this other input information into the LLM 30, thereby acquiring another routing information.

[0043] On the other hand, when the process proceeds from step S206 to step S208, the providing unit 107 provides the routing information acquired by the acquiring unit 104 to a predetermined destination such as the administrator terminal 20.

[0044] By the processing of steps S204 to S208, the routing information providing device 100 repeatedly creates other input information and acquires other routing information until the current network QoS value satisfies the QoS requirements.

[0045] In this way, the routing information providing device 100 according to this embodiment uses the LLM 30 to calculate routing information (routing solution) for avoiding congestion. Furthermore, the routing information providing device 100 uses the calculation unit 105 to evaluate the routing information output by the LLM 30, and if the QoS requirements are not satisfied, it repeats the procedure of instructing the LLM 30 to generate different routing information. This allows the routing information providing device 100 to easily provide routing information for resolving congestion while suppressing a decrease in QoS when congestion occurs in the network.

[0046] <Hardware Configuration> The routing information providing device 100 has, for example, the hardware configuration of a computer 300 as shown in Fig. 3. Alternatively, the routing information providing device 100 is realized by a plurality of computers 300.

[0047] 3 is a diagram showing the hardware configuration of a computer according to this embodiment. In the example of FIG. 3, a computer 300 includes a processor 301, a memory 302, a storage device 303, a communication device 304, an input device 305, an output device 306, and a bus B.

[0048] The processor 301 is, for example, an arithmetic unit such as a CPU (Central Processing Unit) that executes predetermined programs to realize various functions. The memory 302 is a storage medium readable by the computer 300 and includes, for example, a RAM (Random Access Memory) and a ROM (Read Only Memory). The storage device 303 is a computer-readable storage medium and may include, for example, a HDD (Hard Disk Drive), an SSD (Solid State Drive), various optical disks, and magneto-optical disks.

[0049] The communication device 304 includes one or more pieces of hardware (communication devices) for communicating with other devices via a wireless or wired network. The input device 305 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 306 is an output device (e.g., a display, a speaker, an LED lamp, etc.) that outputs to the outside.

[0050] The bus B is commonly connected to the above components and transmits, for example, address signals, data signals, and various control signals. The processor 301 is not limited to a CPU, and may be, for example, a DSP (Digital Signal Processor), a PLD (Programmable Logic Device), or an FPGA (Field Programmable Gate Array).

[0051] (Supplementary Note) The routing information providing device 100 in this embodiment is not limited to being realized by a dedicated device, but may also be realized by a general-purpose computer. In this case, a program for realizing this function may be recorded on a computer-readable recording medium, and the program recorded on this recording medium may be read into a computer system and executed to realize the function. Note that the term "computer system" here includes hardware such as an OS and peripheral devices.

[0052] Furthermore, "computer-readable recording media" includes various storage devices such as portable media such as flexible disks, optical magnetic disks, ROMs, and CD-ROMs, as well as hard disks built into computer systems. Furthermore, "computer-readable recording media" may also include devices that dynamically store programs for a short period of time, such as communication lines when transmitting programs via networks such as the Internet or communication lines such as telephone lines, and devices that store programs for a certain period of time, such as volatile memory within computer systems that serve as servers or clients in such cases.

[0053] Furthermore, the above program may be one that realizes part of the above-mentioned functions, or may be one that can realize the above-mentioned functions in combination with a program already recorded in a computer system, or may be one that is realized using hardware such as a PLD (Programmable Logic Device) or FPGA (Field Programmable Gate Array).

[0054] <Effects of the embodiment> According to the routing information providing device 100 of this embodiment, when congestion occurs in a network, it becomes possible to easily provide routing information for resolving the congestion. Furthermore, according to the routing information providing device 100, it is possible to quickly obtain routing information that is suited to the current network usage status of the user and the target network.

[0055] Furthermore, if the routing information provided by the routing information providing device 100 obtains routing information from the LLM 30 and does not satisfy the QoS requirements, the routing information providing device 100 repeats the process of obtaining the routing information again, thereby providing routing information for resolving congestion while suppressing degradation of QoS.

[0056] In addition, this embodiment makes it easy to provide appropriate routing information by focusing on the fact that although it is difficult to calculate an appropriate routing solution that avoids congestion, it is very easy to verify a routing solution that avoids congestion by calculating a QoS value.

[0057] In addition, in this embodiment, by using a large-scale language model (LLM30), it becomes possible to utilize natural language expressions, and to process information regarding congestion situations and user usage situations that are difficult to express using numerical data.

[0058] Summary of Embodiments This specification discloses at least the routing information providing device, routing information providing system, routing information providing method, and program of the following paragraphs: (1) A routing information providing device comprising: a creation unit that creates input information based on congestion information, network information, and QoS requirements; and an acquisition unit that inputs the input information to the creation unit that generates routing information for avoiding the congestion based on the input information and acquires the routing information. (2) The routing information providing device of paragraph 1, wherein the creation unit is a large-scale language model, and the creation unit creates the input information that is a prompt that instructs the large-scale language model to generate the routing information. (3) The routing information providing device of paragraph 1 or 2, comprising: a calculation unit that calculates a current QoS value of the network based on information about the network, user usage of the network, and the routing information acquired by the acquisition unit. (4) The routing information providing device according to claim 3, wherein, when the current QoS value of the network does not satisfy the QoS requirements, the creation unit creates different input information that instructs the large-scale language model to generate different routing information, and the acquisition unit inputs the different input information into the large-scale language model to acquire the different routing information. (5) The routing information providing device according to claim 4, wherein the creation of the different input information and the acquisition of the different routing information are repeatedly executed until the current QoS value of the network satisfies the QoS requirements. (6) The routing information providing device according to any of claims 1 to 5, wherein the information related to the network includes configuration information of the network, bandwidth setting information of the network, and / or information on routing setting candidates. (7) The information providing device according to any of claims 1 to 6, wherein the information related to the congestion includes information representing, in text format, information on the location where the congestion occurred, time-series data of the communication quality of the network, and / or time-series data of user usage status of the network.(Clause 8) A program, or a storage medium storing a program, that causes a computer to execute: a process of creating input information based on information about network congestion, information about the network, and QoS requirements for the network; and a process of inputting the input information into a generation unit that generates routing information for avoiding network congestion based on the input information, to obtain the routing information. (Clause 9) A routing information providing system comprising: a creation unit that creates input information based on information about congestion, information about the network, and QoS requirements, and an acquisition unit that inputs the input information into a generation unit that generates routing information for avoiding network congestion based on the input information, to obtain the routing information. (Clause 10) A routing information providing method in which a computer executes: a process of creating input information based on information about network congestion, information about the network, and QoS requirements for the network; and a process of inputting the input information into a generation unit that generates routing information for avoiding network congestion based on the input information, to obtain the routing information.

[0059] Although the present embodiment has been described above, the present invention is not limited to such a specific embodiment, and various modifications and changes are possible within the scope of the gist of the present invention described in the claims.

[0060] REFERENCE SIGNS LIST 1 Routing information providing system 10 Monitoring system 20 Administrator terminal 30 LLM (large scale language model, generation unit) 102 Information aggregation unit 103 Creation unit 104 Acquisition unit 105 Calculation unit 106 Determination unit 107 Provision unit 108 Storage unit

Claims

1. A routing information providing device comprising: a creation unit that creates input information based on congestion information, network information, and QoS requirements; and an acquisition unit that inputs the input information into the creation unit that generates routing information for avoiding the congestion based on the input information and acquires the routing information.

2. The routing information providing device according to claim 1, wherein the generation unit is a large-scale language model, and the creation unit creates the input information which is a prompt that instructs the large-scale language model to generate the routing information.

3. A routing information providing device as described in claim 2, further comprising a calculation unit that calculates the current QoS value of the network based on information about the network, user usage of the network, and the routing information acquired by the acquisition unit.

4. A routing information providing device as described in claim 3, wherein, if the current QoS value of the network does not satisfy the QoS requirements, the creation unit creates different input information that instructs the large-scale language model to generate different routing information, and the acquisition unit inputs the different input information into the large-scale language model to acquire the different routing information.

5. The routing information providing device according to claim 4, wherein the creation of the other input information and the acquisition of the other routing information are repeatedly executed until the current QoS value of the network satisfies the QoS requirement.

6. A routing information providing device according to claim 1 or 2, wherein the information relating to the network includes configuration information of the network, bandwidth setting information of the network, and / or information on routing setting candidates.

7. An information providing device as described in claim 1 or 2, wherein the information regarding the congestion includes information in text format representing information on the location where the congestion occurred, time series data of the QoS value of the network, and / or time series data of user usage status of the network.

8. A program that causes a computer to execute the following processes: creating input information based on information about network congestion, information about the network, and QoS requirements of the network; and inputting the input information into a generation unit that generates routing information for avoiding network congestion based on the input information, and acquiring the routing information.

Images