Spare resource availability prediction with limited historical data
A system can comprise a memory that stores computer executable components, and a processor that executes the computer executable components stored in the memory. The computer executable components can comprise a correlation component that, based on a first context defining a first spare resource instance to be utilized in a networked system, identifies a second spare resource instance, and an evaluation component that generates an availability score defining prediction of availability of the first spare resource instance based on a second context defining the second spare resource instance, wherein the evaluation component further deploys the first spare resource instance, based on the availability score, in the networked system.
1 . A system for predicting availability of unknown spare resource instances in a networked system, wherein the unknown spare resource instances have lower spare resource instance interruption data available, identifiable and/or offered in comparison to known spare resource instances in the networked system, the system comprising:
a memory that stores computer executable components; and
a processor that executes the computer executable components stored in the memory, wherein the computer executable components comprise:
a correlation component configured to, based on a first context defining an unknown spare resource instance to be utilized in the networked system, identify a known spare resource instance of the known spare resource instances in the networked system;
an evaluation component configured to generate an availability score defining prediction of availability of the unknown spare resource instance based on a second context that defines the known spare resource instance,
wherein the correlation component is further configured to generate a correlation score defining a correlation between the unknown spare resource instance and the known spare resource instance,
wherein the availability score is generated to identify availability of the unknown spare resource instance based at least in part on the correlation score defining the correlation between the unknown spare resource instance and the known spare resource instance, and
wherein the evaluation component is further configured to deploy the unknown spare resource instance or the known spare resource instances for utilization in the networked system based on the availability score.
2 . The system of claim 1 , wherein an identifiable second amount of second contexts, including the second context, defining the known spare resource instance is greater than an identifiable first amount of first contexts, including the first context, defining the unknown spare resource instance.
3 . The system of claim 1 , wherein the correlation component further identifies the known spare resource instance based on the correlation score being higher than a further correlation score of the unknown spare resource instance with a further known spare resource instance, and wherein the further correlation score defines a correlation between the unknown spare resource instance and the further known spare resource instance.
4 . The system of claim 1 , further comprising:
an analytical model that aggregates contexts, including the first context and the second context, of a plurality of spare resource instances, including the unknown spare resource instance and the known spare resource instance, wherein the analytical model generates a profile correlating the plurality of spare resource instances to one another, and wherein the known spare resource is identified using the profile.
5 . The system of claim 1 , wherein the evaluation component further generates the availability score based on interruption data defining a past interruption of a further known spare resource instance that is related to the known spare resource instance, and wherein the past interruption comprises a rescinding of the further known spare resource instance by a service provider of the known spare resource instance.
6 . The system of claim 5 , wherein the past interruption data comprises a quantity of time elapsed since occurrence of the past interruption of the further known spare resource instance.
7 . The system of claim 1 , further comprising a determination component that identifies the first context defining the unknown spare resource instance, wherein the unknown spare resource instance is defined as a new spare resource instance or a replacement of an already deployed spare resource instance.
8 . The system of claim 1 , wherein the prediction of availability comprises an indication of a predicted amount of uninterrupted time before rescinding of the unknown spare resource instance by a service provider of the unknown spare resource instance.
9 . A computer-implemented method for predicting availability of unknown spare resource instances in a networked system, wherein the unknown spare resource instances have lower spare resource instance interruption data available, identifiable and/or offered in comparison to known spare resource instances in the networked system, the method comprising:
identifying, by a processor, a first context defining an unknown spare resource instance to be utilized in the networked system;
identifying a known spare resource instance of the known spare resource instances in the networked system based on the first context defining the unknown spare resource instance to be utilized in the networked system;
evaluating, by the processor, the identified known spare resource instance to determine a second context defining the known spare resource instance of the known spare resource instances in the networked system;
generating, by the processor, a correlation score defining a correlation between the unknown spare resource instance and the identified known spare resource instance;
generating, by the processor, an availability score for predicting availability of the unknown spare resource instance based on the second context defining the identified known spare resource instance, wherein the correlation score is employed to generate the availability score to predict availability of the unknown spare resource instance;
determining based on the availability score, by the processor, whether to deploy the unknown spare resource instance or the identified known spare resource in the networked system, wherein the availability score is generated to identify availability of the unknown spare resource instance based at least in part on the correlation score defining the correlation between the unknown spare resource instance and the known spare resource instance; and
deploying the unknown spare resource instance or the known spare resource instances for utilization in the networked system based on the availability score.
10 . The computer-implemented method of claim 9 , wherein an identifiable second amount of second contexts, including the second context, defining the known spare resource instance is greater than an identifiable first amount of first contexts, including the first context, defining the unknown spare resource instance.
11 . The computer-implemented method of claim 9 , further comprising identifying, by the processor, the known spare resource instance based on the correlation score being higher than a further correlation score of the unknown spare resource instance with a further known spare resource instance, wherein the further correlation score defines a correlation between the unknown spare resource instance and the further known spare resource instance.
12 . The computer-implemented method of claim 9 , further comprising:
aggregating, by the processor, contexts, including the first context and the second context, of a plurality of spare resource instances, including the unknown spare resource instance and the known spare resource instance;
generating, by the processor, a profile correlating the plurality of spare resource instances to one another; and
identifying, by the processor, the known spare resource instance, using the profile.
13 . The computer-implemented method of claim 9 , further comprising generating, by the processor, the availability score based on interruption data defining a past interruption of a further known spare resource instance that is related to the known spare resource instance, wherein the past interruption comprises a rescinding of the further known spare resource instance by a service provider of the known spare resource instance, and wherein the past interruption data comprises a quantity of time elapsed since occurrence of the past interruption of the further known spare resource instance.
14 . The method of claim 9 , wherein the prediction of availability comprises an indication of a predicted amount of uninterrupted time before rescinding of the unknown spare resource instance by a service provider of the unknown spare resource instance.
15 . A non-transitory computer readable medium including program instructions embodied therewith, the program instructions executable by a processor to cause the processor to execute operations facilitating a process for predicting availability of unknown spare resource instances in a networked system, wherein the unknown spare resource instances have lower spare resource instance interruption data available, identifiable and/or offered in comparison to known spare resource instances in the networked system, the operations comprising:
identifying, by a processor, a first context defining an unknown spare resource instance to be utilized in the networked system;
identifying a known spare resource instance of the known spare resource instances in the networked system based on the first context defining the unknown spare resource instance to be utilized in the networked system;
evaluating, by the processor, the identified known spare resource instance to determine a second context defining the known spare resource instance of the known spare resource instances in the networked system;
generating, by the processor, a correlation score defining a correlation between the unknown spare resource instance and the identified known spare resource instance;
generating, by the processor, an availability score for predicting availability of the unknown spare resource instance to be utilized in the networked system based on the second context defining the identified known spare resource instance, wherein the availability score is generated based on the correlation score defining a correlation between the unknown spare resource instance and the identified known spare resource instance, wherein the availability score is generated to identify availability of the unknown spare resource instance based at least in part on the correlation score;
determining based on the availability score, by the processor, whether to deploy the unknown spare resource instance or the identified known spare resource instance in the networked system; and
deploying the unknown spare resource instance or the known spare resource instances for utilization in the networked system based on the availability score.
16 . The non-transitory computer readable medium of claim 15 , wherein an identifiable second amount of second contexts, including the second context, defining the known spare resource instance is greater than an identifiable first amount of first contexts, including the first context, defining the unknown spare resource instance.
17 . The non-transitory computer readable medium of claim 15 , wherein the operations further comprise:
identifying, by the processor, the known spare resource instance based on the correlation score being higher than a further correlation score of the unknown spare resource instance with a further known spare resource instance, wherein the further correlation score defines a correlation between the unknown spare resource instance and the further known spare resource instance.
18 . The non-transitory computer readable medium of claim 15 , wherein the operations further comprise:
aggregating, by the processor, contexts, including the context, of a plurality of spare resource instances, including the unknown spare resource instance and the known spare resource instance;
generating, by the processor, a profile correlating the plurality of spare resource instances to one another; and
identifying, by the processor, the second spare resource instance, using the profile.
19 . The non-transitory computer readable medium of claim 15 , wherein the operations further comprise:
generating, by the processor, the availability score based on interruption data defining a past interruption of known spare resource instance that is related to the known spare resource instance, wherein the past interruption comprises a rescinding of the known spare resource instance by a service provider of the known spare resource instance, and wherein the past interruption data comprises a quantity of time elapsed since occurrence of the past interruption of the known spare resource instance.
20 . The non-transitory computer readable medium of claim 15 , wherein the prediction of availability comprises an indication of a predicted amount of uninterrupted time before rescinding of the unknown spare resource instance by a service provider of the unknown spare resource instance.