US 12,470,454 B2
On-demand unikernel for application protection
Leigh Griffin, Waterford (IE); Luigi Mario Zuccarelli, Milan (IT); and Sherine Khoury, Puteaux (FR)
Assigned to Red Hat, Inc., Raleigh, NC (US)
Filed by Red Hat, Inc., Raleigh, NC (US)
Filed on Sep. 15, 2023, as Appl. No. 18/369,074.
Prior Publication US 2025/0097095 A1, Mar. 20, 2025
Int. Cl. H04L 41/0604 (2022.01); H04L 41/0663 (2022.01); H04L 45/28 (2022.01)
CPC H04L 41/0627 (2013.01) [H04L 41/0663 (2013.01); H04L 45/28 (2013.01)] 17 Claims
OG exemplary drawing
 
1. A method comprising:
for each of a set of applications identified as critical to a functioning of a computing environment, generating a corresponding unikernel, wherein the unikernel includes code of the application and kernel functionality;
in response to determining that an interference event is affecting a first application of the set of applications, determining that the interference event is unsustainable based on memory allocation collision errors within a period of time exceeding a threshold, wherein the interference event includes interference with the first application utilizing system resources by a second application, wherein the first application is a safety application of a control system associated with the computing environment, wherein the computing environment is a vehicle control system of a vehicle;
adjusting one or more of a set of thresholds associated with a frequency of a type of interference event and a set of thresholds associated with a magnitude of the type of interference event based on historical driving patterns of the vehicle and current driving conditions of the vehicle;
in response to determining that the interference event is unsustainable, initiating, by a processing device, the unikernel corresponding to the first application; and
performing a failover from the first application to the unikernel corresponding to the first application.