Long-duration time series operational analytics
AssigneeTeradata US, Inc.
InventorsRobert Goettge, Birendra Kumar Sahu
Life of this patent
Estimating the term…
Abstract
Various embodiments herein each include at least one of systems, methods, and software for producing operational analytics that summarize fine-grain time scale behavior over long time durations. Some such embodiments are targeted toward understanding operationally meaningful behavior of complex dynamic systems that are often only apparent at fine-grain time scales. Such behavior occurs rarely and/or only for short durations so the analytics of some embodiments cover long time durations.
Claims17
A method comprising:
- transforming time series data into operationally meaningful categories which results in multiple categorical time series, the transforming including transforming of event data into time series data;
- detecting temporal patterns in the categorical time series data;
- recording detected temporal patterns as discrete time events for subsequent analysis;
- applying categorical time series analysis to the previously generated time series to calculate frequencies of occurrence of categorical values and patterns over time;
- deriving measures of association among categorical variates;
- calculating additional analytics from the derived measures of association among the categorical variates;
- transforming mixed numerical and categorical measurements of the time series data into exclusively categorical measurements;
- and
- mapping the time series data to a common time series structure.
A computer-readable storage medium, with instructions stored thereon that are executable by a computer process to perform data processing activities comprising:
- transforming time series data into operationally meaningful categories which results in multiple categorical time series, the transforming including transforming of event data into time series data;
- detecting temporal patterns in the categorical time series data;
- recording detected temporal patterns as discrete time events for subsequent analysis;
- applying categorical time series analysis to the previously generated time series to calculate frequencies of occurrence of categorical values and patterns over time;
- deriving measures of association among categorical variates;
- calculating additional analytics from the derived measures of association among the categorical variates;
- transforming mixed numerical and categorical measurements of the time series data into exclusively categorical measurements;
- and
- mapping the time series data to a common time series structure.
A system comprising:
- at least one computer processor, at least one network interface device, and at least one memory device;
- and
- instructions stored on the at least one memory device that are executable by the at least one computer process to perform data processing activities comprising:
- transforming time series data into operationally meaningful categories which results in multiple categorical time series, the transforming including transforming of event data into time series data;
- detecting temporal patterns in the categorical time series data;
- recording detected temporal patterns as discrete time events for subsequent analysis;
- applying categorical time series analysis to the previously generated time series to calculate frequencies of occurrence of categorical values and patterns over time;
- and
- deriving measures of association among categorical variates;
- and
- calculating additional analytics from the derived measures of association among the categorical variates;
- wherein the time series data is recorded and stored by at least one database management system performance monitoring process;
- and
- transforming the time series data into operationally meaningful categories includes:
- transforming mixed numerical and categorical measurements of the time series data into exclusively categorical measurements;
- and
- mapping the time series data to a common time series structure.
Dependencies are inferred from the claim text.