Model and system for reasoning with N-step lookahead in policy-based system management
First Claim
1. A processor-based method of managing a hardware storage system, comprising:
- detecting an alarm condition relating to a performance goal of the hardware storage system;
invoking an N-step lookahead engine for simulating operation of the hardware storage system if there are multiple candidate actions that could be taken by the hardware storage system for eliminating the detected alarm condition;
generating N simulated storage system states for each of the candidate actions based on a current system state of the hardware storage system, each simulated storage system state comprising at least one of a storage system resource utilization, a storage system event and a storage system workload characteristic;
selecting a candidate action that generates a simulated storage system state that optimizes a stability of the simulated storage system based on eliminating the detected alarm condition, a prerequisite of the hardware storage system and a transient cost of invoking the selected candidate action, the transient cost being based on an overhead cost of the selected candidate action on resources of the hardware storage system; and
outputting the selected candidate action to the hardware storage system for managing the hardware storage system based on the detected alarm condition.
1 Assignment
0 Petitions
Accused Products
Abstract
When an alarm condition relating to a performance goal of a storage system is detected, a storage management system invokes an N-step lookahead engine for simulating operation of the storage system when there are multiple actions that could be taken by the storage system for eliminating the alarm condition. The N-step lookahead engine generates N possible system states based on a current state of the storage system. The N possible states are based on a cost model of each of the multiple actions. Each cost model is based on an action, a behavior implication of the action, a resource implication of the action and a transient cost of the action. An action is selected that generates a system state that optimizes the stability, a prerequisite and a transient cost of invoking the selected action.
37 Citations
20 Claims
-
1. A processor-based method of managing a hardware storage system, comprising:
-
detecting an alarm condition relating to a performance goal of the hardware storage system; invoking an N-step lookahead engine for simulating operation of the hardware storage system if there are multiple candidate actions that could be taken by the hardware storage system for eliminating the detected alarm condition; generating N simulated storage system states for each of the candidate actions based on a current system state of the hardware storage system, each simulated storage system state comprising at least one of a storage system resource utilization, a storage system event and a storage system workload characteristic; selecting a candidate action that generates a simulated storage system state that optimizes a stability of the simulated storage system based on eliminating the detected alarm condition, a prerequisite of the hardware storage system and a transient cost of invoking the selected candidate action, the transient cost being based on an overhead cost of the selected candidate action on resources of the hardware storage system; and outputting the selected candidate action to the hardware storage system for managing the hardware storage system based on the detected alarm condition. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A processor-based system for managing a hardware storage system, comprising:
-
sensors detecting an alarm condition relating to a performance goal of the hardware storage system; and a storage management system invoking an N-step lookahead engine for simulating operation of the hardware storage system if there are multiple candidate actions that could be taken by the hardware storage system for eliminating the detected alarm condition, the N-step lookahead engine generating N simulated storage system states for each of the candidate actions based on a current system state of the hardware storage system, each simulated storage system state comprising at least one of a storage system resource utilization, a storage system event and a storage system workload characteristic, the storage management system selecting a candidate action that generates a simulated storage system state that optimizes a stability of the simulated storage system based on eliminating the detected alarm condition, a prerequisite of the hardware storage system and a transient cost of invoking the selected candidate action, the transient cost being based on an overhead cost of the selected candidate action on resources of the hardware storage system. - View Dependent Claims (12, 13, 14, 15, 16, 17, 18, 19, 20)
-
Specification