System performance degradation mitigation based on different compute resource characteristics

A system for mitigating performance degradation, comprising:

• one or more memories;
• and
• at least one processor each coupled to at least one of the one or more memories and configured to perform operations comprising:
• receiving a first notification indicative of a characteristic of a compute resource of the system, wherein the first notification is of a first type of a plurality of types of notifications, wherein each type of notification of the plurality of types of notifications is indicative of a different characteristic of the same compute resource, wherein the compute resource comprises at least the one or more memories, and wherein the characteristic comprises at least a percentage of time tasks are delayed based on memory allocation time of the one or more memories;
• determining a first level of performance degradation, from a plurality of different levels of performance degradation, of the system based on the first notification utilizing a mapping that maps each type of notification of the plurality of types of notifications to a corresponding level of performance degradation of the plurality of different levels of performance degradation;
• causing, based on the determined first level of performance degradation, a background application executing on the system to be terminated;
• receiving a second notification indicative of the characteristic of the compute resource, wherein the second notification is of a second type of the plurality of types of notifications;
• determining a second level of performance degradation, from the plurality of different levels of performance degradation, of the system based on the second notification utilizing the mapping, wherein the second level of performance degradation is more severe than the first level of performance degradation;
• and
• causing, based on the determined second level of performance degradation, a foreground application executing on the system to be terminated.

A non-transitory computer-readable medium having instructions stored thereon that, when executed by at least one computing device, cause the at least one computing device to perform operations comprising:

• receiving a first notification indicative of a characteristic of a compute resource of the at least one computing device, wherein the first notification is of a first type of a plurality of types of notifications, wherein each type of notification of the plurality of types of notifications is indicative of a different characteristic of the same compute resource, wherein the compute resource comprises at least one or more memories of the at least one computing device, and wherein the characteristic comprises at least a percentage of time tasks are delayed based on memory allocation time of the one or more memories;
• determining a first level of performance degradation, from a plurality of different levels of performance degradation, of the at least one computing device based on the first notification utilizing a mapping that maps each type of notification of the plurality of types of notifications to a corresponding level of performance degradation of the plurality of different levels of performance degradation;
• causing, based on the determined first level of performance degradation, a background application executing on the at least one computing device to be terminated;
• receiving a second notification indicative of the characteristic of the compute resource, wherein the second notification is of a second type of the plurality of types of notifications;
• determining a second level of performance degradation, from the plurality of different levels of performance degradation, of the at least one computing device based on the second notification utilizing the mapping, wherein the second level of performance degradation is more severe than the first level of performance degradation;
• and
• causing, based on the determined second level of performance degradation, a foreground application executing on the at least one computing device to be terminated.