Enriched high fidelity metrics

A system comprising:

• a data repository storing:
• a plurality of metrics describing operational behavior of a plurality of software programs executing in an enterprise system, wherein each set of the plurality of metrics represents data for a different time period in a plurality of time periods;
• a plurality of events, wherein each of the plurality of events represents a behavior of interest over time in the plurality of software programs, wherein at least one of the plurality of events comprises a fault in a first software program;
• metadata that correlates the plurality of events to at least some of the plurality of metrics over the plurality of time periods;
• a plurality of ingested metrics comprising the plurality of metrics enriched with the metadata;
• a plurality of time slices, wherein each of the plurality of time slices comprises a corresponding data structure for storing the plurality of ingested metrics in the plurality of time periods;
• a plurality of heat maps stored in the plurality of time slices, the plurality of heat maps summarizing operational behavior of the plurality of software programs in ones of the plurality of time periods;
• an application programming interface (API) gateway configured to receive the plurality of metrics;
• an ingestion layer configured to ingest the plurality of metrics to form the plurality of ingested metrics;
• and
• a tumbling window processor configured to process the plurality of ingested metrics and the plurality of events into the plurality of heat maps, sort the plurality of heat maps into the plurality of time slices, and populate the plurality of time slices with the plurality of ingested metrics.

A method comprising:

• receiving a first plurality of metrics describing a first plurality of behaviors of a plurality of applications;
• receiving a second plurality of metrics describing a second plurality of behaviors of an enterprise system in which the plurality of applications are executing, wherein at least one of the first plurality of metrics and the second plurality of metrics indicates a fault in at least one of the plurality of applications;
• ingesting the first plurality of metrics and the second plurality of metrics to generate ingested metrics, wherein ingesting comprises:
• assigning a first cardinality score to first metrics from at least one of the first plurality of metrics and the second plurality of metrics that are related to the fault, and
• assigning a second cardinality score to second metrics from at least one of the first plurality of metrics and the second plurality of metrics that are unrelated to the fault, wherein the first cardinality score is higher than the second cardinality score;
• performing tumbling window processing on the ingested metrics by:
• sorting the first plurality of metrics and the second plurality of metrics into a plurality of time slices, wherein each of the plurality of time slices comprises a corresponding data structure for storing data in different time periods, and
• storing more data in the plurality of time slices for the first metrics having the first cardinality score relative to the second metrics having the second cardinality score;
• and
• displaying the plurality of time slices on a display device.