Model driven configuration of software installation operating states
Product usage may be modeled based on large volume, machine generated product usage data to optimize product pricing and operations. Specifically, the systems and methods described herein may utilize methods with key components to select the maximum number of dimensions that can be modeled based on the number of data points, use a logarithm kernel function to normalize machine data with long-tailed statistical distributions on different numerical scales, compare a large number of candidate models with different candidate dimensions and different structures, and quantify the amount of change and drift in models over time.
1. A computing device, comprising:
one or more hardware processors; and
a non-transitory computer-readable medium having stored thereon instructions that, when executed by the one or more hardware processors, cause the one or more hardware processors to perform operations including:
generating a plurality of candidate models using a machine learning model, wherein the plurality of candidate models predict values for a first set of data points indicating product usage, each data point included in the first set of data points has a first set of dimensions, and each candidate model included in the plurality of candidate models is generated based on the first set of data points using data from a different subset of the first set of dimensions;
generating respective model testing results for each candidate model based on an evaluation of the respective candidate model;
selecting a model from the plurality of candidate models by identifying a model from the plurality of candidate models with a lowest median error based upon the model testing results;
determining, using the selected model, a first quantity of computing resources or operating states of a software installation for a customer based on a time series analysis of a recent quantity of computing resources or operating states of the customer;
determining a second quantity of computing resources or operating states of the software installation based on one or more of the following: static customer data, software performance data, and time series analysis of a quantity of computing resources or operating states of the customer;
determining a third quantity of computing resources or operating states of the software installation based on the first quantity of computing resources or operating states and the second quantity of computing resources or operating states; and
causing the software installation to be executed based on a parameter or operating state of the software installation that has been adjusted according to the third quantity.
2. The computing device of claim 1 , wherein the operations further comprise:
determining to adjust the parameter of the software installation based on a majority of the first quantity of computing resources, the second quantity of computing resources, and the third quantity of computing resources being larger than a current quantity of computing resources.
3. The computing device of claim 2 ,
wherein the operations further comprise:
determining, as voting indicators, which of the first quantity of computing resources, the second quantity of computing resources, are larger than a current quantity of computing resources;
wherein the determining of the third quantity of computing resources or operating states comprises selecting a quantity of computing resources from among the voting indicators.
4. The computing device of claim 1 , wherein:
the determining of the third quantity of computing resources or operating states comprises selecting either the first quantity of computing resources or the second quantity of computing resources.
5. The computing device of claim 1 , wherein:
the determining of the third quantity of computing resources or operating states comprises selecting the larger of the first quantity of computing resources and the second quantity of computing resources.
6. The computing device of claim 1 , wherein the parameter of the software installation for the customer is adjusted based on the first quantity of computing resources being larger than a current quantity of computing resources.
7. The computing device of claim 1 , wherein the determining of the second quantity of computing resources or operating states of the software installation is based on the static customer data.
8. The computing device of claim 1 , wherein the determining of the second quantity of computing resources or operating states of the software installation is based on the software performance data.
9. The computing device of claim 1 , wherein the determining of the second quantity of computing resources or operating states of the software installation is based on the time series analysis of the quantity of computing resources or operating states of the customer.
10. A computer-implemented method, comprising:
generating, by one or more hardware processors, a plurality of candidate models using a machine learning model, wherein the plurality of candidate models predict values for a first set of data points indicating product usage, each data point included in the first set of data points has a first set of dimensions, and each candidate model included in the plurality of candidate models is generated based on the first set of data points using data from a different subset of the first set of dimensions;
generating respective model testing results for each candidate model based on an evaluation of the respective candidate model;
selecting a model from the plurality of candidate models by identifying a model from the plurality of candidate models with a lowest median error based upon the model testing results;
determining, using the selected model, a first quantity of computing resources or operating states of a software installation for a customer based on a second set of data points corresponding to a recent quantity of computing resources or operating states of the customer;
determining a second quantity of computing resources or operating states of the software installation based on one or more of the following: static customer data, software performance data, and time series analysis of a quantity of computing resources or operating states of the customer;
determining a third quantity of computing resources or operating states of the software installation based on the first quantity of computing resources or operating states and the second quantity of computing resources or operating states; and
causing the software installation to be executed based on a parameter or operating state of the software installation that has been adjusted according to the third quantity.
11. The computer-implemented method of claim 10 , further comprising:
determining to adjust the parameter of the software installation based on a majority of the first quantity of computing resources, the second quantity of computing resources, and the third quantity of computing resources being larger than a current quantity of computing resources.
12. The computer-implemented method of claim 11 ,
further comprising:
determining, as voting indicators, which of the first quantity of computing resources, the second quantity of computing resources, are larger than the current quantity of computing resources,
wherein the determining of the third quantity of computing resources or operating states comprises selecting a quantity of computing resources from among the voting indicators.
13. The computer-implemented method of claim 10 , wherein:
the determining of the third quantity of computing resources or operating states comprises selecting either the first quantity of computing resources or the second quantity of computing resources.
14. The computer-implemented method of claim 10 , wherein:
the determining of the third quantity of computing resources or operating states comprises selecting the larger of the first quantity of computing resources and the second quantity of computing resources.
15. The computer-implemented method of claim 10 , wherein the parameter of the software installation for the customer is adjusted based on the first quantity of computing resources being larger than a current quantity of computing resources.
16. The computer-implemented method of claim 10 , wherein the determining of the second quantity of computing resources or operating states of the software installation is based on the static customer data.
17. The computer-implemented method of claim 10 , wherein the determining of the second quantity of computing resources or operating states of the software installation is based on the software performance data.
18. A non-transitory computer-readable medium having stored thereon instructions that, when executed by one or more processors, cause the one or more processors to perform operations including:
generating a plurality of candidate models using a machine learning model, wherein the plurality of candidate models predict values for a first set of data points indicating product usage, each data point included in the first set of data points has a first set of dimensions, and each candidate model included in the plurality of candidate models is generated based on the first set of data points using data from a different subset of the first set of dimensions;
generating respective model testing results for each candidate model based on an evaluation of the respective candidate model;
selecting a model from the plurality of candidate models by identifying a model from the plurality of candidate models with a lowest median error based upon the model testing results;
determining, using the selected model, a first quantity of computing resources or operating states of a software installation for a customer based on a second set of data points corresponding to a recent quantity of computing resources or operating states of the customer;
determining a second quantity of computing resources or operating states of the software installation based on one or more of the following: static customer data, software performance data, and time series analysis of a quantity of computing resources or operating states of the customer;
determining a third quantity of computing resources or operating states of the software installation based on the first quantity of computing resources or operating states and the second quantity of computing resources or operating states; and
causing the software installation to be executed based on adjusting a parameter or operating state of the software installation that has been adjusted according to the third quantity.
19. The non-transitory computer-readable medium of claim 18 , wherein the operations further comprise:
determining to adjust the parameter of the software installation based on a majority of the first quantity of computing resources, the second quantity of computing resources, and the third quantity of computing resources being larger than a current quantity of computing resources.
20. The non-transitory computer-readable medium of claim 19 ,
wherein the operations further comprise:
determining, as voting indicators, which of the first quantity of computing resources, the second quantity of computing resources, are larger than the current quantity of computing resources,
wherein the determining of the third quantity of computing resources or operating states comprises selecting a quantity of computing resources from among the voting indicators.