Systems and methods for encoding knowledge for automated management of software application deployments

A method of encoding knowledge that can be used to automatically detect problems in software application deployments, comprising:

• accessing a source of knowledge describing a problem known to occur in deployments of a particular software application, and which identifies a plurality of conditions associated with the problem;
• generating an encoded representation of the knowledge source according to a predefined knowledge encoding methodology, said encoded representation adapted to be applied automatically by a computer to analyze data representing a current state of a monitored deployment of said software application to detect whether said conditions and said problem exist therein;
• and
• storing the encoded representation in computer storage, in a format in which the encoded representation can be used by the computer to automatically detect the problem in said monitored deployment of said software application;
• wherein the encoded representation of the knowledge source includes logic for automatically generating, in connection with a detection of an occurrence of the problem, a confidence value representing a confidence that the detection is accurate;
• wherein the encoded representation of the knowledge source specifies at least one query for affirmatively causing a collection of new telemetry data from said monitored deployment of the software application to obtain a new value of a time-varying state metric of the deployment when the encoded representation has not been previously used to evaluate said state metric during the detection of a particular one of the conditions, wherein the query is represented in the encoded representation, the encoded representation being configured to be used in combination with at least the value obtained via said query to evaluate whether the problem exists in the deployment.

A method of encoding knowledge that can be used to automatically detect problems in software application deployments, comprising:

• accessing a source of knowledge describing a problem known to occur in deployments of a particular software application, and which identifies a plurality of conditions associated with the problem;
• generating an encoded representation of the knowledge source according to a predefined knowledge encoding methodology, said encoded representation adapted to be applied automatically by a computer to analyze state data of a monitored deployment of said software application to detect whether said conditions and said problem exist therein, wherein the encoded representation can be programmatically used to receive condition confidence values each representing a confidence that a particular one of the conditions is true;
• providing within the encoded representation static significance values for the conditions, such that the significance values can be programmatically used to decline to report the problem when (1) one or more of the conditions are found to be true, (2) the true conditions indicate that the problem is currently present in the monitored deployment, and (3) a condition confidence value of at least one of the true conditions does not exceed the significance value of that particular condition;
• and
• storing the encoded representation in computer storage, in a format in which the encoded representation can be used by the computer to automatically detect the problem in said monitored deployment of said software application.