Predicting and adjusting computer functionality to avoid failures

• US 10,338,994 B1
• Filed: 10/19/2018
• Issued: 07/02/2019
• Est. Priority Date: 02/22/2018
• Status: Active Grant

• Current Assignee: SAS Institute Incorporated
• Original Assignee: SAS Institute Incorporated
• Inventors: Xie, Jingrui, Li, Yue, Chien, Yung-Hsin, Wang, Pu
• Primary Examiner(s): Amoroso, Anthony J
• Application Number:
  US16/165,331
• Time in Patent Office: 256 Days
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/0727 in a storage system, e.g. in a DASD or network based storage system drivers for digital recording or reproducing units G06F3/06; circuits for error detection or correction within digital recording or reproducing units G11B20/18; for distributed storage of data in networks, e.g. transport arrangements for network file system [NFS], storage area networks [SAN] or network attached storage [NAS], H04L67/1097
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/0775 Content or structure details of the error report, e.g. specific table structure, specific error fields
• CPC Class Codes: G06F 11/0793 Remedial or corrective actions recovery from an exception in an instruction pipeline G06F9/3861; by retry G06F11/1402; for recovering from a failure of a protocol instance or entity H04L69/40
• CPC Class Codes: G06F 11/2028 eliminating a faulty processor or activating a spare
• CPC Class Codes: G06F 11/2041 with more than one idle spare processing component
• CPC Class Codes: G06F 11/2048 where the redundant components share neither address space nor persistent storage
• CPC Class Codes: G06F 11/2097 maintaining the standby controller/processing unit updated initialisation or re-synchronisation thereof G06F11/1658 and subgroups
• CPC Class Codes: G06F 11/3006 where the computing system is distributed, e.g. networked systems, clusters, multiprocessor systems multiprogramming arrangements G06F9/46; allocation of resources G06F9/50
• CPC Class Codes: G06F 11/3051 Monitoring arrangements for monitoring the configuration of the computing system or of the computing system component, e.g. monitoring the presence of processing resources, peripherals, I/O links, software programs verification or detection of system hardware configuration G06F11/2247
• CPC Class Codes: G06F 11/3058 Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations thermal management in cooling arrangements of a computing system G06F1/206
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: H04L 41/0816 the condition being an adaptation, e.g. in response to network events
• CPC Class Codes: H04L 41/147 for predicting network behaviour
• CPC Class Codes: H04L 41/149 for prediction of maintenance

Abstract

In some examples, a processing device can receive prediction data representing a prediction. The processing device can also receive files defining abnormal data-point patterns to be identified in the prediction data. The processing device can identify at least one abnormal data-point pattern in the prediction data by executing customizable program-code in the files. The processing device can determine an override process that corresponds to the at least one abnormal data-point pattern in response to identifying the at least one abnormal data-point pattern in the prediction data. The processing device can execute the override process to generate a corrected version of the prediction data. The processing device can then adjust one or more computer parameters based on the corrected version of the prediction data.

Claims

1. A system comprising:

a processing device; and

a memory device including instructions that are executable by the processing device for causing the processing device to;

receive prediction data representing a prediction, wherein the prediction data forms a time series that spans a future time-period;

receive a plurality of files defining abnormal data-point patterns to be identified in the prediction data, wherein each file in the plurality of files includes customizable program-code for identifying a respective abnormal pattern of data-point values in the prediction data;

automatically identify a plurality of abnormal data-point patterns in the prediction data by interpreting and executing the customizable program-code in the plurality of files;

automatically determine a plurality of override processes that correspond to the plurality of abnormal data-point patterns in response to identifying the plurality of abnormal data-point patterns in the prediction data, wherein the plurality of override processes are automatically determined using correlations between the plurality of abnormal data-point patterns and the plurality of override processes, and wherein an override process involves replacing a value of at least one data point in the prediction data with another value that is configured to mitigate an impact of an abnormal data-point pattern on the prediction;

automatically determine that the plurality of override processes are to be applied to the prediction data in a particular order;

automatically generate a corrected version of the prediction data in response to determining the plurality of override processes, wherein the corrected version of the prediction data is generated by executing the plurality of override processes in the particular order; and

automatically adjust one or more computer parameters based on the corrected version of the prediction data.

2. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to determine the particular order based on a configuration file specifying the particular order.

3. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to flag the prediction as potentially erroneous in response to identifying the at last one plurality of abnormal data-point patterns in a historical dataset used to generate the prediction data.

4. The system of claim 1, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically identify the plurality of abnormal data-point patterns in the prediction data by performing only a single pass through the prediction data.

5. The system of claim 1, wherein the plurality of abnormal data-point patterns includes a first abnormal data-point pattern and a second abnormal data-point pattern, and wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to:

automatically determine the second abnormal data-point pattern associated with the prediction data by analyzing characteristics of a historical dataset used to generate the prediction data;

automatically determine that the first abnormal data-point pattern corresponds to a first override-process and the second abnormal data-point pattern corresponds to a second override-process using the correlations; and

automatically generate the corrected version of the prediction data at least in part by executing the second override-process in the particular order with respect to the first override-process.

6. The system of claim 5, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically determine the second abnormal data-point pattern by:

determining that the historical dataset used to generate the prediction data has a series of consecutive data points with values of zero spanning at least an entire season cycle within the historical dataset.

7. The system of claim 5, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically determine the second abnormal data-point pattern by:

determining that an end of the prediction data has consecutive data-points with values of zero;

determining whether any timespan in the historical dataset that was used to generate the prediction data includes at least as many consecutive data-points with values of zero as are at the end of the prediction data; and

detecting the second abnormal data-point pattern in response to determining that there is no timespan in the historical dataset that includes at least as many consecutive data-points with values of zero as are at the end of the prediction data.

8. The system of claim 5, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically determine that the prediction data has the second abnormal data-point pattern by:

generating a historical profile for a timespan of the historical dataset that was used to generate the prediction data, wherein the historical profile is an average of multiple seasons of information within the historical dataset;

determining that the timespan in the historical profile is an active period in which there are no consecutive data-points with values of zero;

analyzing the same timespan in the prediction to determine that the timespan in the prediction data is an inactive period in which there are consecutive data-points with values of zero; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the timespan represents an active period in the historical profile and an inactive period in the prediction data.

9. The system of claim 5, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically determine that the prediction data has the second abnormal data-point pattern by:

determining an average value of the prediction data;

determining another average value of the historical dataset that was used to generate the prediction data;

determining that the average value of the prediction data is different from the other average value of the historical dataset by more than a threshold amount; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the average value of the prediction data is different from the other average value of the historical dataset by more than the threshold amount.

10. The system of claim 5, wherein the memory device further includes instructions that are executable by the processing device for causing the processing device to automatically determine that the prediction data has the second abnormal data-point pattern by:

automatically comparing (i) a data-point pattern during a time period in the prediction data to (ii) another data-point pattern during the time period in the historical dataset that was used to generate the prediction data in order to determine a difference between the data-point patterns; and

automatically determining that the difference between the data-point patterns is indicative of the second abnormal data-point pattern in the prediction data.

11. A non-transitory computer-readable medium comprising program code that is executable by a processing device for causing the processing device to:

receive prediction data representing a prediction, wherein the prediction data forms a time series that spans a future time-period;

receive a plurality of files defining abnormal data-point patterns to be identified in the prediction data, wherein each file in the plurality of files includes customizable program-code for identifying a respective abnormal pattern of data-point values in the prediction;

automatically identify a plurality of abnormal data-point patterns in the prediction data by interpreting and executing the customizable program-code in the plurality of files;

automatically determine a plurality of override processes that correspond to the plurality of abnormal data-point patterns in response to identifying the plurality of abnormal data-point patterns in the prediction data, wherein the plurality of override processes are automatically determined using correlations between the plurality of abnormal data-point patterns and the plurality of override processes, and wherein an override process involves replacing a value of at least one data point in the prediction data with another value that is configured to mitigate an impact of an abnormal data-point pattern on the prediction;

automatically determine that the plurality of override processes are to be applied to the prediction data in a particular order;

automatically generate a corrected version of the prediction data in response to determining the plurality of override processes, wherein the corrected version of the prediction data is generated by executing the plurality of override processes in the particular order; and

automatically adjust one or more computer parameters based on the corrected version of the prediction data.

12. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to determine the particular order based on a configuration file specifying the particular order.

13. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to flag the prediction data as potentially erroneous in response to identifying the plurality of abnormal data-point patterns in a historical dataset that was used to generate the prediction data.

14. The non-transitory computer-readable medium of claim 11, further comprising program code that is executable by the processing device for causing the processing device to automatically identify the plurality of abnormal data-point patterns in the prediction data by performing only a single pass through the prediction data.

15. The non-transitory computer-readable medium of claim 11, wherein the plurality of abnormal data-point patterns includes a first abnormal data-point pattern and a second abnormal data-point pattern, and further comprising program code that is executable by the processing device for causing the processing device to:

automatically determine the second abnormal data-point pattern associated with the prediction data by analyzing characteristics of a historical dataset used to generate the prediction data;

automatically determine that the first abnormal data-point pattern corresponds to a first override-process and the second abnormal data-point pattern corresponds to a second override-process using the correlations; and

automatically generate the corrected version of the prediction data at least in part by executing the second override-process in the particular order with respect to the first override-process.

16. The non-transitory computer-readable medium of claim 15, further comprising program code that is executable by the processing device for causing the processing device to automatically determine the second abnormal data-point pattern by:

determining that the historical dataset that was used to generate the prediction data has a series of consecutive data points with values of zero spanning at least an entire season cycle within the historical dataset.

17. The non-transitory computer-readable medium of claim 15, further comprising program code that is executable by the processing device for causing the processing device to automatically determine the second abnormal data-point pattern by:

determining that an end of the prediction data has consecutive data-points with values of zero;

determining whether any timespan in the historical dataset that was used to generate the prediction data includes at least as many consecutive data-points with values of zero as are at the end of the prediction data; and

detecting the second abnormal data-point pattern in response to determining that there is no timespan in the historical dataset that includes at least as many consecutive data-points with values of zero as are at the end of the prediction data.

18. The non-transitory computer-readable medium of claim 15, further comprising program code that is executable by the processing device for causing the processing device to automatically determine that the prediction has the second abnormal data-point pattern by:

generating a historical profile for a timespan of the historical dataset that was used to generate the prediction data, wherein the historical profile is an average of multiple seasons of information within the historical dataset;

determining that the timespan in the historical profile is an active period in which there are no consecutive data-points with values of zero;

analyzing the same timespan in the prediction data to determine that the timespan in the prediction data is an inactive period in which there are consecutive data-points with values of zero; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the timespan represents an active period in the historical profile and an inactive period in the prediction data.

19. The non-transitory computer-readable medium of claim 15, further comprising program code that is executable by the processing device for causing the processing device to automatically determine that the prediction data has the second abnormal data-point pattern by:

determining an average value of the prediction data;

determining another average value of the historical dataset that was used to generate the prediction data;

determining that the average value of the prediction data is different from the other average value of the historical dataset by more than a threshold amount; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the average value of the prediction data is different from the other average value of the historical dataset by more than the threshold amount.

20. The non-transitory computer-readable medium of claim 15, further comprising program code that is executable by the processing device for causing the processing device to automatically determine that the prediction has the second abnormal data-point pattern by:

automatically comparing (i) a data-point pattern during a time period in the prediction data to (ii) another data-point pattern during the time period in the historical dataset that was used to generate the prediction data in order to determine a difference between the data-point patterns; and

automatically determining that the difference between the data-point patterns is indicative of the second abnormal data-point pattern in the prediction data.

21. A method comprising:

receiving, by a processing device, prediction data representing a prediction, wherein the prediction data forms a time series that spans a future time-period;

receiving, by the processing device, a plurality of files defining abnormal data-point patterns to be identified in the prediction data, wherein each file in the plurality of files includes customizable program-code for identifying a respective abnormal pattern of data-point values in the prediction data;

automatically identifying, by the processing device, a plurality of abnormal data-point patterns in the prediction data by interpreting and executing the customizable program-code in the plurality of files;

automatically determining, by the processing device, a plurality of override processes that correspond to the plurality of abnormal data-point patterns in response to identifying the plurality of abnormal data-point patterns in the prediction data, wherein the plurality of override processes are automatically determined using correlations between the plurality of abnormal data-point patterns and the plurality of override processes, and wherein an override process involves replacing a value of at least one data point in the prediction data with another value that is configured to mitigate an impact of an abnormal data-point pattern on the prediction;

automatically determining, by the processing device, that the plurality of override processes are to be applied to the prediction data in a particular order;

automatically generating, by the processing device, a corrected version of the prediction data in response to determining the plurality of override processes, wherein the corrected version of the prediction data is generated by executing the plurality of override processes in the particular order; and

automatically adjusting, by the processing device, one or more computer parameters based on the corrected version of the prediction data.

22. The method of claim 21, further comprising determining the particular order based on a configuration file specifying the particular order.

23. The method of claim 21, further comprising flagging the prediction as potentially erroneous in response to identifying the plurality of abnormal data-point patterns in a historical dataset used to generate the prediction data.

24. The method of claim 21, further comprising automatically identifying the plurality of abnormal data-point patterns in the prediction data by performing only a single pass through the prediction data.

25. The method of claim 21, wherein the plurality of abnormal data-point patterns includes a first abnormal data-point pattern and a second abnormal data-point pattern, and further comprising:

automatically determining the second abnormal data-point pattern associated with the prediction data by analyzing characteristics of a historical dataset used to generate the prediction data;

automatically determining that the first abnormal data-point pattern corresponds to a first override-process and the second abnormal data-point pattern corresponds to a second override-process using the correlations; and

automatically generating the corrected version of the prediction data at least in part by executing the second override-process in the particular order with respect to the first override-process.

26. The method of claim 25, further comprising automatically determining the second abnormal data-point pattern by:

determining that the historical dataset that was used to generate the prediction data has a series of consecutive data points with values of zero spanning at least an entire season cycle within the historical dataset.

27. The method of claim 25, further comprising automatically determining the second abnormal data-point pattern by:

determining that an end of the prediction data has consecutive data-points with values of zero;

determining whether any timespan in the historical dataset that was used to generate the prediction data includes at least as many consecutive data-points with values of zero as are at the end of the prediction data; and

detecting the second abnormal data-point pattern in response to determining that there is no timespan in the historical dataset that includes at least as many consecutive data-points with values of zero as are at the end of the prediction data.

28. The method of claim 25, further comprising automatically determining that the prediction data has the second abnormal data-point pattern by:

generating a historical profile for a timespan of the historical dataset that was used to generate the prediction data, wherein the historical profile is an average of multiple seasons of information within the historical dataset;

determining that the timespan in the historical profile is an active period in which there are no consecutive data-points with values of zero;

analyzing the same timespan in the prediction data to determine that the timespan in the prediction data is an inactive period in which there are consecutive data-points with values of zero; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the timespan represents an active period in the historical profile and an inactive period in the prediction.

29. The method of claim 25, further comprising automatically determining that the prediction has the second abnormal data-point pattern by:

determining an average value of the prediction data;

determining another average value of the historical dataset that was used to generate the prediction data;

determining that the average value of the prediction data is different from the other average value of the historical dataset by more than a threshold amount; and

determining that the prediction data has the second abnormal data-point pattern in response to determining that the average value of the prediction data is different from the other average value of the historical dataset by more than the threshold amount.

30. The method of claim 25, further comprising automatically determining that the prediction has the second abnormal data-point pattern by:

automatically comparing (i) a data-point pattern during a time period in the prediction data to (ii) another data-point pattern during the time period in the historical dataset that was used to generate the prediction data in order to determine a difference between the data-point patterns; and

automatically determining that the difference between the data-point patterns is indicative of the second abnormal data-point pattern in the prediction data.