公开(公告)号：US11863398B2

公开(公告)日：2024-01-02

申请号：US17484349

申请日：2021-09-24

Applicant: Google LLC

Inventor： Rajeev Koodli ,  Mazin Gilbert

IPC: H04L41/16 ,  H04L67/10 ,  G06N5/04 ,  G06N20/00

CPC classification number: H04L41/16 ,  G06N5/04 ,  G06N20/00 ,  H04L67/10

Abstract: Aspects of the disclosure are directed to a central management plane (CMP) of one or more processors for regulating streams of data from each of a number of network nodes of a distributed network. The one or more processors can train and deploy machine learning models across the network nodes, and the CMP can generate policies for each network node. The generated policies specify how a network node is to transmit data to the platform for further training or retraining of the deployed machine learning models. The CMP generates the policies using metric data characterizing each network node and respective streams of input data, and are generated based on a number of objectives, including model output quality of the deployed models, and operational cost to transmit and process streams of data across the distributed network.

公开(公告)号：US20230267089A1

公开(公告)日：2023-08-24

申请号：US18112740

申请日：2023-02-22

Applicant: Google LLC

Inventor： Santanu Dasgupta ,  Durgaprasad V. Ayyadevara ,  Bor Chan ,  Prashant R. Chandra ,  Bok Knun Randolph Chung ,  Max Kamenetsky ,  Rajeev Koodli ,  Shahin Valoth

IPC: G06F13/38 ,  G06F13/12

CPC classification number: G06F13/385 ,  G06F13/122 ,  G06F2213/0038

Abstract: The present disclosure provides a compute platform architecture for virtualized and cloud native network functions. The architecture uses a reduced instruction set computer-based general purpose processor along with multiple special purpose accelerators and an integrated network interface card. As such, the architecture can accommodate multiple hundreds of gigabits of input/output.

公开(公告)号：US11425602B2

公开(公告)日：2022-08-23

申请号：US16854140

申请日：2020-04-21

Applicant: Google LLC

Inventor： Rajeev Koodli

IPC: H04W28/08 ,  G06N20/00 ,  H04L101/654 ,  H04W8/08 ,  H04W48/18

Abstract: A method for service selection in mobile networks includes receiving from a control plane, a request for a user plane instance. The user plane instance is configured to perform packet processing for a user equipment during a communication session. The method also includes identifying a plurality of user plane instance candidates associated with a base station in communication with the user equipment. The plurality of user plane instance candidates is configurable by the control plane. For each user plane instance candidate, the method includes determining one or more selection parameters corresponding to a subset of key performance indicators for the base station. The method further includes selecting one of the plurality of user plane instance candidates to fulfill the request for the user plane instance from the control plane based on the one or more selection parameters determined for each of the plurality of user plane instance candidates.

公开(公告)号：US20230388197A1

公开(公告)日：2023-11-30

申请号：US18231832

申请日：2023-08-09

Applicant: Google LLC

Inventor： Rajeev Koodli ,  Mazin Gilbert

IPC: H04L41/16 ,  H04L67/10 ,  G06N5/04 ,  G06N20/00

CPC classification number: H04L41/16 ,  H04L67/10 ,  G06N5/04 ,  G06N20/00

Abstract: Aspects of the disclosure are directed to a central management plane (CMP) of one or more processors for regulating streams of data from each of a number of network nodes of a distributed network. The one or more processors can train and deploy machine learning models across the network nodes, and the CMP can generate policies for each network node. The generated policies specify how a network node is to transmit data to the platform for further training or retraining of the deployed machine learning models. The CMP generates the policies using metric data characterizing each network node and respective streams of input data, and are generated based on a number of objectives, including model output quality of the deployed models, and operational cost to transmit and process streams of data across the distributed network.

公开(公告)号：US20230078246A1

公开(公告)日：2023-03-16

申请号：US17484349

申请日：2021-09-24

Applicant: Google LLC

Inventor： Rajeev Koodli ,  Mazin Gilbert

IPC: H04L12/24 ,  G06N20/00 ,  G06N5/04 ,  H04L29/08

Abstract: Aspects of the disclosure are directed to a central management plane (CMP) of one or more processors for regulating streams of data from each of a number of network nodes of a distributed network. The one or more processors can train and deploy machine learning models across the network nodes, and the CMP can generate policies for each network node. The generated policies specify how a network node is to transmit data to the platform for further training or retraining of the deployed machine learning models. The CMP generates the policies using metric data characterizing each network node and respective streams of input data, and are generated based on a number of objectives, including model output quality of the deployed models, and operational cost to transmit and process streams of data across the distributed network.

公开(公告)号：US12199835B2

公开(公告)日：2025-01-14

申请号：US18231832

申请日：2023-08-09

Applicant: Google LLC

Inventor： Rajeev Koodli ,  Mazin Gilbert

IPC: H04L41/16 ,  G06N5/04 ,  G06N20/00 ,  H04L67/10

Abstract: Aspects of the disclosure are directed to a central management plane (CMP) of one or more processors for regulating streams of data from each of a number of network nodes of a distributed network. The one or more processors can train and deploy machine learning models across the network nodes, and the CMP can generate policies for each network node. The generated policies specify how a network node is to transmit data to the platform for further training or retraining of the deployed machine learning models. The CMP generates the policies using metric data characterizing each network node and respective streams of input data, and are generated based on a number of objectives, including model output quality of the deployed models, and operational cost to transmit and process streams of data across the distributed network.

公开(公告)号：US20210329496A1

公开(公告)日：2021-10-21

申请号：US16854140

申请日：2020-04-21

Applicant: Google LLC

Inventor： Rajeev Koodli

IPC: H04W28/08 ,  H04L29/12 ,  H04W48/18 ,  H04W8/08 ,  G06N20/00

Abstract: A method for service selection in mobile networks includes receiving from a control plane, a request for a user plane instance. The user plane instance is configured to perform packet processing for a user equipment during a communication session. The method also includes identifying a plurality of user plane instance candidates associated with a base station in communication with the user equipment. The plurality of user plane instance candidates is configurable by the control plane. For each user plane instance candidate, the method includes determining one or more selection parameters corresponding to a subset of key performance indicators for the base station. The method further includes selecting one of the plurality of user plane instance candidates to fulfill the request for the user plane instance from the control plane based on the one or more selection parameters determined for each of the plurality of user plane instance candidates.