Decreasing downtime of computer systems using predictive detection
A master processor may retrieve historical and real time machine and human data related to computer system health. The master processor may utilize machine learning and artificial intelligence to predict potential computer malfunctions. The master processor may output notifications regarding the potential computer malfunctions in order to prevent the computer malfunctions from occurring.
1. A method, comprising:
retrieving, by a processor, machine data from a machine data source;
retrieving, by the processor, incident report data describing one or more computer malfunctions from an incident report data source;
retrieving, by the processor, a change record representing a scheduled change to a configuration of a computer system, the change record specifying when the scheduled change is scheduled to occur, a number of phases in which the scheduled change will be implemented, and one or more environments in which the scheduled change has been tested;
retrieving, by the processor, an algorithm from a model library;
determining, by the processor using the algorithm and based at least in part on the scheduled change described by the change record, the machine data, and the incident report data, a likelihood that the change record will cause a computer malfunction in the computer system;
determining, by the processor using the algorithm and based at least in part on the machine data and the incident report data, an estimated cost associated with the computer malfunction; and
generating, by the processor, a change record evaluation, the change record evaluation comprising the likelihood that the change record will cause the computer malfunction and the estimated cost associated with the computer malfunction.
2. The method of claim 1 , wherein the machine data comprises historical data and real time data.
3. The method of claim 1 , further comprising correlating, by the processor and using the machine data, a first spike in computing resource consumption by a first application with a second spike in computing resource consumption by a second application, the first spike corresponding to a first increase of at least three standard deviations over average variability for the computing resource consumption by the first application, and the second spike corresponding to a second increase of at least three standard deviations over average variability for the computing resource consumption by the second application.
4. The method of claim 1 , wherein the change record comprises a software upgrade and a time of upgrade.
5. The method of claim 1 , further comprising recalibrating, by the processor, the algorithm based at least in part on feedback from a subject matter expert.
6. The method of claim 5 , wherein the problem record identifies a previous problem, and wherein the incident record identifies a current problem.
7. The method of claim 1 , further comprising selecting the algorithm from a plurality of algorithms stored in the model library based at least in part on an accuracy of the algorithm.
8. A system comprising:
a processor; and
a tangible, non-transitory memory configured to communicate with the processor, the tangible, non-transitory memory having instructions stored thereon that, in response to execution by the processor, cause the processor to perform operations comprising:
retrieving, by the processor, machine data from a machine data source;
retrieving, by the processor, incident report data describing one or more computer malfunctions from an incident report data source;
retrieving, by the processor, a change record representing a scheduled change to a configuration of a computer system, the change record specifying when the scheduled change is scheduled to occur, a number of phases in which the scheduled change will be implemented, and one or more environments in which the scheduled change has been tested;
retrieving, by the processor, an algorithm from a model library;
determining, by the processor using the algorithm and based at least in part on the scheduled change described by the change record, the machine data, and the incident report data, a likelihood that the change record will cause a computer malfunction in the computer system;
determining, by the processor using the algorithm and based at least in part on the machine data and the incident report data, an estimated cost associated with the computer malfunction; and
generating, by the processor, a change record evaluation, the change record evaluation comprising the likelihood that the change record will cause the computer malfunction and the estimated cost associated with the computer malfunction.
9. The system of claim 8 , wherein the machine data comprises historical data and real time data.
10. The system of claim 8 , the operations further comprising correlating, by the processor and using the machine data, a first spike in computing resource consumption by a first application with a second spike in computing resource consumption by a second application, the first spike corresponding to a first increase of at least three standard deviations over average variability for the computing resource consumption by the first application, and the second spike corresponding to a second increase of at least three standard deviations over average variability for the computing resource consumption by the second application.
11. The system of claim 8 , wherein the change record comprises a software upgrade and a time of upgrade.
12. The system of claim 8 , the operations further comprising recalibrating, by the processor, the algorithm based at least in part on feedback from a subject matter expert.
13. The system of claim 12 , wherein the problem record identifies a previous problem, and wherein the incident record identifies a current problem.
14. The system of claim 8 , the operations further comprising selecting the algorithm from a plurality of algorithms stored in the model library based at least in part on an accuracy of the algorithm.
15. An article of manufacture including a non-transitory, tangible computer readable storage medium having instructions stored thereon that, in response to execution by a computer based system, cause the computer based system to perform operations comprising:
retrieving, by a processor, machine data from a machine data source;
retrieving, by the processor, incident report data describing one or more computer malfunctions from an incident report data source;
retrieving, by the processor, a change record representing a scheduled change to a configuration of a computer system, the change record specifying when the scheduled change is scheduled to occur, a number of phases in which the scheduled change will be implemented, and one or more environments in which the scheduled change has been tested;
retrieving, by the processor, an algorithm from a model library;
determining, by the processor using the algorithm and based at least in part on the scheduled change described by the change record, the machine data, and the incident report data, a likelihood that the change record will cause a computer malfunction in the computer system;
determining, by the processor using the algorithm and based at least in part on the machine data and the incident report data, an estimated cost associated with the computer malfunction; and
generating, by the processor, a change record evaluation, the change record evaluation comprising the likelihood that the change record will cause the computer malfunction and the estimated cost associated with the computer malfunction.
16. The article of manufacture of claim 15 , wherein the machine data comprises historical data and real time data.
17. The article of manufacture of claim 15 , the operations further comprising correlating, by the processor and using the machine data, a first spike in computing resource consumption by a first application with a second spike in computing resource consumption by a second application, the first spike corresponding to a first increase of at least three standard deviations over average variability for the computing resource consumption by the first application, and the second spike corresponding to a second increase of at least three standard deviations over average variability for the computing resource consumption by the second application.
18. The article of manufacture of claim 15 , wherein the change record comprises a software upgrade and a time of upgrade.
19. The article of manufacture of claim 15 , the operations further comprising recalibrating, by the processor, the algorithm based at least in part on feedback from a subject matter expert.
20. The article of manufacture of claim 19 , wherein the problem record identifies a previous problem, and wherein the incident record identifies a current problem.