Machine learning to predict quality-of-service needs in an operational data management system
Operational data in a distributed processing system is managed by monitoring a workload of the system to establish a current assessment of operational data movement between data sources and data targets, receiving historical information on previous data movement including previous instances of movement resulting in a compromise of one or more quality-of-service criteria, determining from the current assessment and historical information that upcoming operational data actions will not meet a particular quality-of-service criterion, and responsively applying a data management optimization infrastructure (data backplane services) adapted to advance the particular quality-of-service criterion according to definitions for the data sources and data targets. The operational outcome is predicted using a cognitive system trained with historical information including historical operational factors correlated with historical operational outcomes relative to the quality-of-service criteria.
1. A method of assessing resource needs to meet quality-of-service criteria for a distributed processing system comprising:
training a cognitive system with historical information on previous operation of the distributed processing system, the historical information including historical operational factors of the distributed processing system correlated with historical operational outcomes relative to the quality-of-service criteria;
receiving current operational factors of the distributed processing system; and
using the cognitive system to generate a predicted operational outcome of the distributed processing system based on the current operational factors, the predicted operational outcome providing an indication that a particular one of the quality-of-service criteria will not be met, wherein the predicted operational outcome assigns values based on confidence values generated by the cognitive system to indicate the likelihood of quality-of-service failure in quantitative form.
2. The method of claim 1 wherein the historical operational factors include at least temporal information reflecting peak usages times, resource availability, customer information, transaction load, and network traffic.
3. The method of claim 1 wherein the historical operational outcomes are based on data movement parameters.
4. The method of claim 1 , wherein:
the distributed processing system provides a fraud detection solution; and
the quality-of-service criteria include at least resource allocations, data integrity specifications, and service uptime.
5. The method of claim 1 further comprising transmitting the predicted operational outcome to a data movement optimization module which invokes one or more data backplane services as needed to advance the particular quality-of-service criteria.
6. The method of claim 5 wherein the data backplane services provide feedback which is used to update the historical operational factors.
7. The method of claim 5 wherein:
the predicted operational outcome provides an indication that multiple quality-of-service criteria will not be met with respective likelihood values; and
the data movement optimization module prioritizes the data backplane services according to the likelihood values.
8. A computer system comprising:
one or more processors which process program instructions;
a memory device connected to said one or more processors; and
program instructions residing in said memory device for assessing resource needs to meet quality-of-service criteria for a distributed processing system by training a cognitive system with historical information on previous operation of the distributed processing system, the historical information including historical operational factors of the distributed processing system correlated with historical operational outcomes relative to the quality-of-service criteria, receiving current operational factors of the distributed processing system, and using the cognitive system to generate a predicted operational outcome of the distributed processing system based on the current operational factors, the predicted operational outcome providing an indication that a particular one of the quality-of-service criteria will not be met, wherein the predicted operational outcome assigns values based on confidence values generated by the cognitive system to indicate the likelihood of quality-of-service failure in quantitative form.
9. The computer system of claim 8 wherein the historical operational factors include at least temporal information reflecting peak usages times, resource availability, customer information, transaction load, and network traffic.
10. The computer system of claim 8 wherein the historical operational outcomes are based on data movement parameters.
11. The computer system of claim 8 wherein:
the distributed processing system provides a fraud detection solution; and
the quality-of-service criteria include at least resource allocations, data integrity specifications, and service uptime.
12. The computer system of claim 8 further comprising transmitting the predicted operational outcome to a data movement optimization module which invokes one or more data backplane services as needed to advance the particular quality-of-service criteria.
13. The computer system of claim 12 wherein:
the predicted operational outcome provides an indication that multiple quality-of-service criteria will not be met with respective likelihood values; and
the data movement optimization module prioritizes the data backplane services according to the likelihood values.
14. The computer system of claim 12 wherein the data backplane services provide feedback which is used to update the historical operational factors.
15. A computer program product comprising:
a computer readable storage medium; and
program instructions residing in said storage medium for assessing resource needs to meet quality-of-service criteria for a distributed processing system by training a cognitive system with historical information on previous operation of the distributed processing system, the historical information including historical operational factors of the distributed processing system correlated with historical operational outcomes relative to the quality-of-service criteria, receiving current operational factors of the distributed processing system, and using the cognitive system to generate a predicted operational outcome of the distributed processing system based on the current operational factors, the predicted operational outcome providing an indication that a particular one of the quality-of-service criteria will not be met, wherein the predicted operational outcome assigns values based on confidence values generated by the cognitive system to indicate the likelihood of quality-of-service failure in quantitative form.
16. The computer program product of claim 15 wherein the historical operational factors include at least temporal information reflecting peak usages times, resource availability, customer information, transaction load, and network traffic.
17. The computer program product of claim 15 wherein the historical operational outcomes are based on data movement parameters.
18. The computer program product of claim 15 wherein:
the distributed processing system provides a fraud detection solution; and
the quality-of-service criteria include at least resource allocations, data integrity specifications, and service uptime.
19. The computer program product of claim 15 wherein further comprising transmitting the predicted operational outcome to a data movement optimization module which invokes one or more data backplane services as needed to advance the particular quality-of-service criteria.
20. The computer program product of claim 19 wherein:
the predicted operational outcome provides an indication that multiple quality-of-service criteria will not be met with respective likelihood values; and
the data movement optimization module prioritizes the data backplane services according to the likelihood values.