Rakuten Mobile, Inc.

Rakuten Symphony, Inc.

Rakuten Symphony Singapore Pte. Ltd.

Shinya Kita

Bharath Rathinam

Mohit Luthra

Jithin Chathankandath

Puneet Devadiga

Rahul Atri

Abhishek Sharma

Rajat Singh

Tareq Amin

Rajasi Ahuja

Mihir Pathak

Amey Sandesh Wadekar

Ajay Bukan

An inventory database and an active inventory each store inventory data indicating a current status of an influence relationship between elements included in the communication system. An E2EO module detects occurrence of an event in a specific element included in the communication system. The E2EO module identifies at least one other element having an influence relationship with the specific element based on the inventory data. The E2EO module identifies a cause of the event based on a status of the at least one other element.

Cause specifying system and cause specifying method

Provided are a router estimation system and a router estimation method which enable accurate estimation of a router in network slices that is a cause of a silent failure. A policy manager () extracts, when it is determined that a plurality of function elements have deteriorated in performance in slice communication, a plurality of function elements that are similar to one another in terms of shifts in performance in the slice communication from the plurality of function elements determined to have deteriorated in performance. The policy manager () identifies, for each of the plurality of extracted function elements, a group of routers located on a route of the slice communication based on router group data. The policy manager () estimates at least one router included in every router group identified for the each of the plurality of extracted function elements to be a router that is a cause of the deterioration of performance of the plurality of extracted function elements.

Estimation of router that is cause of silent failures

A slice manager module selects two network nodes out of network nodes belonging to the network slice and determines the two network nodes as the pair. The slice manager module determines an IP address relating to the network slice for each of the network nodes of the pair. An SDN controller determines, as a part or all of a communication route of the network slice, a communication route between the determined IP addresses. The SDN controller re-determines the communication route between the IP addresses in response to a fact that communication quality of the determined communication route satisfies a predetermined condition.

Communication route determination system and communication route determination method

Provided are a performance index value calculation system and a performance index value calculation method which are capable of reducing time and labor for an operation of a communication system. A monitoring management module identifies, based on inventory data indicating a current status of a configuration of an element group included in a communication system and on calculation logic data indicating a condition to be satisfied by an element used for calculating a desired performance index value, a monitoring item for the element, and a calculation logic for calculating the desired performance index value, an element that satisfies the condition from the element group indicated by the inventory data. The monitoring management module calculates the desired performance index value by applying the calculation logic indicated by the calculation logic data to a performance index value of the monitoring item for the identified element.

Performance index value calculation system and performance index value calculation method

An E2EO module determines, in response to determination that a new element linked to an already installed element included in the communication system is to be added to the communication system, a location at which the new element is to be constructed based on new topology data including the new element and a current location of the already installed element indicated by location data.

Location determination system and location determination method

Provided are a router estimation system and a router estimation method which enable accurate estimation of a router in network slices that is a cause of a silent failure. A policy manager () identifies, when a plurality of function elements are determined to have deteriorated in performance in slice communication, for each of the plurality of function elements determined to have deteriorated in performance, a group of routers located on a route of the slice communication, based on router group data. The policy manager () estimates at least one router included in every router group identified for the each of the plurality of function elements determined to have deteriorated in performance to be a router that is a cause of the deterioration of performance of the function element.

A network operating system receives a bundle file transmitted from a vendor terminal of a vendor providing a network service, the bundle file including first data defining a functional unit group that achieves the network service and second data defining a monitoring policy for the network service. The NOS constructs the functional unit group based on the first data of the bundle file when the network service is purchased by a purchaser. The NOS executes a monitoring process on the functional unit group based on information on the functional unit group to be constructed and the second data of the bundle file.

Computer system and network service construction method

A network operating system (NOS) acquires, for a plurality of network functions that achieve a network slice constructed on a mobile wireless communication network, information on an operation of each network function which is measured by one or more apparatus configured to provide at least a part of the plurality of network functions. The NOS derives an evaluation index value of each of the plurality of network functions based on the information on the operation of each of the plurality of network functions, and derives an evaluation index value of the network slice based on the evaluation index values of the plurality of network functions.

Computer system and network slice management method

Provided are a router estimation system and a router estimation method which enable accurate estimation of a router that is a cause of silent failures in network slices. When it is determined that a degree of correlation increase linked to a pair of a first slice communication and a second slice communication satisfies a given condition, a policy manager () estimates at least one router included in both of a first router group, which is a group of routers identified based on router group data as routers located on a route of the first slice communication, and a second router group, which is a group of routers identified based on the router group data as routers located on a route of the second slice communication, as a router that is a cause of deterioration of performance of a first function element and a second function element.

An E2EO module identifies a monitoring item for which a user is to be provided with a monitoring result. The E2EO module determines, based on the identified monitoring item, an exclusive-use element from an element group including at least one of a software element group that achieves a network service or a hardware element group in which the software element group is implemented. The E2EO module, a configuration management module, and a container management module construct the software element group that achieves the network service in a configuration in which at least the exclusive-use element is exclusively used by the user.

Network service build system and network service build method

To accurately infer a cause of a trouble occurring in a network. A network system acquires first information indicating a status of the network (S), selects, from among a plurality of cause inference models provided for each type of a cause of a failure occurring in the network, each of the cause inference models being associated with a model selection condition, a cause inference model corresponding to the model selection condition satisfied by the acquired first information, the plurality of cause inference models being trained based on training data including input data that includes second information indicating the status of the network in a time period different from the time period of the first information and ground truth data indicating a cause of a trouble (S), acquires an output obtained when the first information indicating the status of the network is input to the selected cause inference model (S), and executes a process for responding to a trouble that has occurred in the network based on the acquired output of the selected cause inference model (S).

Cause inference regarding network trouble

A communication system is appropriately scaled out. An AI predicts, based on a performance index value before a given reference time point, a network load at a plurality of prediction time points included in a prediction period from the given reference time point until a predetermined time after the given reference time point. A policy manager determines, for each of the plurality of prediction time points, whether a predicted value of the network load at the prediction time point exceeds a threshold value, which increases as a length of time from the reference time point until the prediction time point increases. The policy manager, a life cycle manager, a container manager, and a configuration manager execute scale-out of an element included in the communication system when it is determined that the predicted value exceeds the threshold value at a part or all of the plurality of prediction time points.

Executing appropriate scale-out of an element included in a communication system

It is enabled that network load prediction for executing scale-out of an element included in a communication system is started at an appropriate timing. A policy manager () determines whether network load prediction is required based on performance index value data and a determination criterion which is based on a usage amount of computer resources. An AI () starts network load prediction in response to a determination that the network load prediction is required. After the network load prediction is started, the policy manager () determines whether scale-out is required based on a network load prediction result. The policy manager (), a life cycle manager (), a container manager (), and a configuration manager () execute scale-out of the element included in the communication system in response to a determination that the scale-out is required.

Controlling start timing of network load prediction

To appropriately determine a trouble occurring in a network. A network system acquires an output obtained when input data including an index acquired from a target network is input to a cause inference model. The cause inference model is one of a plurality of cause inference models. The plurality of cause inference models respectively correspond to a plurality of groups into which a plurality of networks have been classified and are trained based on training data including input data including an index acquired for each corresponding group and ground truth data indicating a cause of a trouble. The input cause inference model corresponds to a group to which the target network of the plurality of networks belongs. The network system executes a process for responding to a trouble that has occurred in the target network based on the output of the trouble cause inference model for the target network.

Provided are a performance index value calculation system and a performance index value calculation method which are capable of reducing time and labor for an operation of a communication system. A monitoring management module identifies a specific type of element included in a communication system based on inventory data indicating a current status of a link between elements included in the communication system and on calculation logic data indicating a calculation logic for calculating a performance index value of the specific type of element based on performance index values of an element group linked to the element. The monitoring management module identifies the performance index value of each of a plurality of elements included in the element group linked to the identified specific type of element. The monitoring management module calculates the performance index value of the specific type of element by applying the calculation logic indicated by the calculation logic data to the performance index value identified for each of the plurality of elements.

Growing community of inventors

Tokyo, Japan

Shinya Kita