Schema-based computer system health monitoring

A computer-readable medium having stored thereon a data structure, comprising:

• a first data field representing a collection object associated with a resource, the collection object including collection properties that specify information for collecting data from the resource;
• a second data field representing a threshold object associated with the collection object, the threshold object including threshold properties that specify information for evaluating the data collected from the resource;
• and
• a third data field representing an action object, the action object defining action properties that specify throttling data and at least one action to perform when at least one of the threshold properties is achieved and the throttling data indicates that the action should be performed.

In a computer system, a method for monitoring a plurality of system resources, comprising:

• grouping datagroup objects into a hierarchy, at least some of the datagroup objects representing resources;
• attaching at least one action object to the hierarchy, each action object including property information that determines performance of an action when a status event reaches the action object;
• associating a collection object with one of the datagroup objects representing a resource, the collection object including information for collecting data from the resource associated with the collection object;
• associating a threshold object with the collection object, the threshold object including properties specifying information for evaluating the collected data and for sending a status event to towards the top of the hierarchy when the data evaluation indicates a threshold violation state;
• collecting data from the resource via the collection object;
• evaluating the data collected from the resource via the threshold object, determining a threshold violation state and sending a status event to towards the top of the hierarchy;
• and
• receiving the status event at an action object, and in response, selectively performing an action based on the property information of the action object.

In a computing environment, a system comprising:

• a plurality of datagroup objects arranged in a hierarchy, at least some of the datagroup objects associated with a computing resource;
• a plurality of collection objects, each collection object corresponding to a resource and having code that when executed collects data related to its corresponding resource;
• a plurality of threshold objects, each threshold object corresponding to a collection object and having code that when executed evaluates the data collected by its corresponding collection object against threshold property data to determine whether a threshold violation exists, and generates a status event when a threshold violation exists, the status event sent towards the top of the hierarchy;
• and
• a plurality of action objects, each action object placed in the hierarchy to receive status events, each action object having code that when executed determines performance of an action based on action property data in the action object.