公开(公告)号：EP3933584A1

公开(公告)日：2022-01-05

申请号：EP20183264.9

申请日：2020-06-30

Applicant: ABB Schweiz AG

Inventor： EIDENBENZ, Raphael ,  PIGNOLET, Yvonne-Anne ,  VUKOVIC, Ognjen ,  MOGA, Alexandru ,  KOZHAYA, David ,  BIRKE, Robert ,  KANDE, Mallikarjun

IPC: G06F9/50 ,  G06F11/30

Abstract: A method for allocating fog applications in a fog network with a plurality of fog nodes is suggested. The method comprises: - providing an application model; - providing a fog network image that reflects the properties of the fog network; - performing a resource usage test of the application model on the fog network image, and receiving resource requirements for the fog network; - creating a concrete application model based on the application model, wherein the concrete application model contains the received resource requirements; and - performing allocation of the concrete application model on one or more of the fog nodes of the fog network.

公开(公告)号：EP3933583A1

公开(公告)日：2022-01-05

申请号：EP20183245.8

申请日：2020-06-30

Applicant: ABB Schweiz AG

Inventor： BIRKE, Robert ,  EIDENBENZ, Raphael ,  PIGNOLET, Yvonne-Anne ,  MOGA, Alexandru ,  RAMASWAMY, Srini

IPC: G06F9/50

Abstract: A method for adjusting machine learning (ML) models in a system comprising a plurality of devices is suggested. The method comprises: - providing a system comprising a plurality of devices, wherein the devices have computational resource capacities; - providing one or more machine learning (ML) tasks; - providing a repository of ML models for the one or more tasks, wherein a plurality of the ML models of a single task solve the same task with different computational resources requirements and different quality metrics; - selecting a device of the plurality of devices of the system to execute a task, wherein the selected device has available computational resource capacities; and - selecting, from the repository of ML models of the task to be executed, one of the ML models, wherein the computational resources requirements of the selected ML model do not exceed the available computational resource capacities of the selected device. Systems configured to perform the methods as disclosed herein are also suggested.