US 12,292,860 B2
Optimal downtime suggestion for hybrid it landscapes
Resmi K S, Palakkad (IN); and Akilandeswari V, Bangalore (IN)
Filed by SAP SE, Walldorf (DE)
Filed on Oct. 28, 2022, as Appl. No. 17/975,959.
Prior Publication US 2024/0143558 A1, May 2, 2024
Int. Cl. G06F 16/00 (2019.01); G06F 16/21 (2019.01); G06F 16/25 (2019.01)
CPC G06F 16/21 (2019.01) [G06F 16/256 (2019.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method, the method comprising:
receiving a request for a suggested downtime of a specified system within a hybrid information technology (IT) landscape, the hybrid IT landscape including at least one on premise system and at least one cloud-based system;
determining, based on information regarding one or more executable processes depending on the specified system to implement the one or more executable processes, at least one on premise contributor to a downtime for the hybrid IT landscape and at least one cloud-based contributor to the downtime for the hybrid IT landscape, the at least one on premise contributor and the at least one cloud-based contributor each being a contributing factor to the downtime for the at least one on premise system and the at least one cloud-based system, respectively and each of the at least one on premise contributor and the at least one cloud-based contributor to a downtime for the hybrid IT landscape having a rank value associated with each of the at least one on premise contributor and the at least one cloud-based contributor;
determining, based on their respective associated rank value, an ordered ranking of each of the at least one on premise contributor and the at least one cloud-based contributor, the ordered ranking providing a relative weight associated with each of the at least one on premise contributor and the at least one cloud-based contributor to a downtime for the hybrid IT landscape;
generating, by a pretrained machine learning model based on an input to the pretrained machine learning model including a combination of the at least one on premise contributor and the at least one cloud-based contributor and their respective associated ordered ranking, one or more downtime time slots;
presenting, in reply to the request and based on the generated downtime time slots, the suggested downtime for the specified system;
receiving, in reply to the presentation of the suggested downtime for the specified system, an indication of a user-selected downtime for the specified system;
in an instance the indication of the user-selected downtime for the specified system is one of the downtime time slots generated by the machine learning model, updating the machine learning model and adjusting the associated rank value for the at least one on premise contributor and the at least one cloud-based contributor based on the user-selected downtime;
in an instance the indication of the user-selected downtime for the specified system is exclusive of one of the downtime time slots generated by the machine learning model, updating the machine learning model and adjusting the associated rank value for the at least one on premise contributor and the at least one cloud-based contributor, wherein the adjusting of the ordered ranking is performed in response to the receiving of the indication exclusive of one of the downtime time slots generated by the machine learning model exceeds a threshold value; and
generating, based on a second input including a combination of at least one on premise contributor and at least one cloud-based contributor and their respective adjusted associated rank value to the updated machine learning model, a second one or more downtime time slots.