A method for monitoring the health of a compute resource includes determining a baseline for the amount of time required for the resource to complete a specific task, monitoring the resource to measure the amount of time actually used by the resource to complete the task in a given instance, comparing the baseline time to the measured time, and determining if the resource is operating within acceptable tolerances based on the comparison. An alert may be created or other remedial action may be taken if the resource is not operating within acceptable tolerances.