Tenant-side detection, classification, and mitigation of noisy-neighbor-induced performance degradation
Embodiments relate to tenant-side detection and mitigation of performance degradation resulting from interference generated by a noisy neighbor in a distributed computing environment. A first machine-learning model such as a k-means nearest neighbor classifier is operated by a tenant to detect an anomaly with a computer system emulator resulting from a co-located noisy neighbor. A second machine-learning model such as a multi-class classifier is operated by the tenant to identify a contended resource associated with the anomaly. A corresponding trigger signal is generated and provided to trigger various mitigation responses, including an application/framework-specific mitigation strategy (e.g., triggered approximations in application/framework performance, best-efforts paths, run-time changes, etc.), load-balancing, scaling out, updates to a scheduler to avoid impacted nodes, and the like. In this manner, a tenant can detect, classify, and mitigate performance degradation resulting from a noisy neighbor.
1. A computer-implemented method comprising:
causing an application or framework of a first tenant of a cloud computing environment to run inside one or more computer system emulators on one or more nodes in the cloud computing environment;
causing a machine-learning model controlled by the first tenant to classify, based on performance metrics of the application or framework, a contended resource of the cloud computing environment associated with interference from a second tenant of the cloud computing environment; and
triggering, by an operation controlled by the first tenant, a mitigation of the interference.
2. The computer-implemented method of claim 1 , further comprising storing the machine-learning model on a master node that coordinate s distributed computations of the application or framework in the cloud computing environment.
3. The computer-implemented method of claim 1 , further comprising co-locating the machine-learning model with at least one of the one or more computer system emulators on the one or more nodes.
4. The computer-implemented method of claim 1 , wherein the one or more computer system emulators comprise one or more software containers, the method further comprising packaging the machine-learning model in one of the one or more software containers for the application or framework.
5. The computer-implemented method of claim 1 , wherein the mitigation of the interference comprises at least one of a tenant-triggered load-balancing or a tenant-triggered scaling out.
6. The computer-implemented method of claim 1 , wherein the mitigation of the interference comprises an application-specific or framework-specific mitigation strategy specific to the application or framework.
7. The computer-implemented method of claim 6 , wherein the application-specific mitigation strategy comprises at least one of an automatically triggered approximation associated with operation of the application or an automatically triggered change to a run-time operation of the application.
8. The computer-implemented method of claim 1 , further comprising causing another machine-learning model controlled by the first tenant to detect the interference based on the performance metrics.
9. One or more computer storage media storing computer-useable instructions, that, when executed by one or more computing devices, cause the one or more computing devices to perform operations comprising:
causing an application or framework of a first tenant of a cloud computing environment to run inside one or more computer system emulators on one or more nodes in the cloud computing environment;
causing a machine-learning model controlled by the first tenant to classify, based on performance metrics of the application or framework, a contended resource of the cloud computing environment associated with interference from a second tenant of the cloud computing environment; and
triggering, based on the contended resource and by an operation controlled by the first tenant, a mitigation of the interference.
10. The one or more computer storage media of claim 9 , the operations further comprising storing the machine-learning model on a master node that coordinates distributed computations of the application or framework in the cloud computing environment.
11. The one or more computer storage media of claim 9 , the operations further comprising co-locating the machine-learning model with at least one of the one or more computer system emulators on the one or more nodes.
12. The one or more computer storage media of claim 9 , wherein the one or more computer system emulators comprise one or more software containers, the machine-learning model packaged in one of the one or more software containers for the application or framework.
13. The one or more computer storage media of claim 9 , wherein the mitigation of the interference comprises at least one of a tenant-triggered load-balancing or a tenant-triggered scaling out.
14. The one or more computer storage media of claim 9 , wherein the mitigation of the interference comprises an application-specific or framework-specific mitigation strategy specific to the application or framework.
15. The one or more computer storage media of claim 14 , wherein the application-specific mitigation strategy comprises at least one of an automatically triggered approximation associated with operation of the application or an automatically triggered change to a run-time operation of the application.
16. A computer system comprising one or more hardware processors and one or more computer storage media storing computer-useable instructions, that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations comprising:
causing an application or framework of a first tenant of a cloud computing environment to run inside one or more computer system emulators on one or more nodes in the cloud computing environment;
causing a machine-learning model controlled by the first tenant to classify, based on performance metrics of the application or framework, a contended resource of the cloud computing environment associated with interference from a second tenant of the cloud computing environment; and
triggering, based on the contended resource and by an operation controlled by the first tenant, a mitigation of the interference.
17. The computer system of claim 16 , the operations further comprising storing the machine-learning model on a master node that coordinate s distributed computations of the application or framework in the cloud computing environment.
18. The computer system of claim 16 , the operations further comprising co-locating the machine-learning model with at least one of the one or more computer system emulators on the one or more nodes.
19. The computer system of claim 16 , wherein the one or more computer system emulators comprise one or more software containers, the machine-learning model packaged in one of the one or more software containers for the application or framework.
20. The computer system of claim 16 , wherein the mitigation of the interference comprises at least one of a tenant-triggered load-balancing or a tenant-triggered scaling out.