US 12,080,324 B2
Data tape media quality validation and action recommendation
Jitesh Bakul Jhatakia, Superior, CO (US); Robert Olin Wyman, Berthoud, CO (US); Frank Patrick Abbott, Jr., Brighton, CO (US); Carl William Luehr, Erie, CO (US); Cathleen Susan Wharton, Louisville, CO (US); John Mitchell Black, III, Loveland, CO (US); Anthony Joseph Braun, Lafayette, CO (US); Scott Adrian Ellett, Broomfield, CO (US); and George Edward Noble, Boulder, CO (US)
Assigned to Oracle International Corporation, Redwood Shores, CA (US)
Filed by Oracle International Corporation, Redwood Shores, CA (US)
Filed on Jun. 15, 2022, as Appl. No. 17/841,574.
Claims priority of provisional application 63/217,040, filed on Jun. 30, 2021.
Claims priority of provisional application 63/217,036, filed on Jun. 30, 2021.
Claims priority of provisional application 63/217,032, filed on Jun. 30, 2021.
Prior Publication US 2023/0005511 A1, Jan. 5, 2023
Int. Cl. G11B 27/36 (2006.01); G06N 20/00 (2019.01)
CPC G11B 27/36 (2013.01) [G06N 20/00 (2019.01)] 18 Claims
OG exemplary drawing
 
1. A non-transitory computer readable medium comprising instructions which, when executed by one or more hardware processors, causes performance of operations comprising:
training a machine learning model to generate action recommendations for data tape cartridges in a data tape system at least by:
obtaining from a data repository system a training data set comprising:
data tape attribute data for the data tape cartridges corresponding to historical exchanges between the data tape cartridges and media drives, wherein the data tape attribute data includes at least:
metadata message attribute data for historical metadata messages corresponding to the historical exchanges between the data tape cartridges and the media drives; and
quality information about the data tape cartridges; and
recommended actions for the data tape cartridges corresponding to the historical metadata messages;
applying a machine learning algorithm to the training data set to train the machine learning model to generate the recommendations;
obtaining, in real-time, from the data tape system, for a data tape cartridge, at least one metadata message based on a first set of sensed data associated with an interaction of the data tape cartridge with a first media drive;
obtaining a data tape quality value based on a second set of sensed data associated with the data tape cartridge, the second set of sensed data based on additional interactions of the data tape cartridge with one or more media drives, including the first media drive; and
applying the machine learning model to (1) first attribute data from the at least one metadata message, and (2) the data tape quality value, to generate a first action recommendation for the data tape cartridge.