US 12,131,146 B2
Maintenance time window prediction for installing updates to a compute node
Mantej Singh Gill, New Delhi (IN); Dhamodhran Sathyanarayanamurthy, Bangalore Karnataka (IN); and Arun Mahendran, Bangalore Karnataka (IN)
Assigned to Hewlett Packard Enterprise Development LP, Spring, TX (US)
Filed by HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP, Spring, TX (US)
Filed on Dec. 23, 2022, as Appl. No. 18/146,096.
Prior Publication US 2024/0211234 A1, Jun. 27, 2024
Int. Cl. G06F 8/65 (2018.01)
CPC G06F 8/65 (2013.01) 20 Claims
OG exemplary drawing
 
1. A system comprising:
at least one computing device including at least one processor configured to:
receive a data payload of update metadata for each update component of a plurality of update components to be installed on a plurality of compute nodes of a system, wherein the update metadata comprises a type of update component, a size of update component, and a type of device to be updated;
determine compute node metadata for each compute node of the plurality of compute nodes to be updated;
determine a predicted update duration for each update component of the plurality of update components by feeding the compute node metadata and the update metadata as inputs into a duration prediction model comprising a function that relates compute node features and update features as independent variables to predict update durations for the plurality of update components as outputs;
aggregate the predicted update duration for each update component of the plurality of update components to generate an aggregated predicted update duration for the plurality of update components; and
generate and format return payload data with the aggregated predicted update duration for the plurality of update components.