Autonomic control of a distributed computing system in accordance with a hierarchical model

• US 7,680,799 B2
• Filed: 03/07/2005
• Issued: 03/16/2010
• Est. Priority Date: 01/31/2005
• Status: Active Grant

• Current Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Original Assignee: Computer Associates Think Inc. (Broadcom, Inc.)
• Inventors: Forney, Brian C., Engquist, James D., Jackson, Jerry R., Murrish, Randall L., Emeis, Mark W., Collins, Doreen E.
• Primary Examiner(s): Vital; Pierre M
• Assistant Examiner(s): Obisesan; Augustine
• Application Number:
  US11/074,291
• Publication Number:
  US 20060173856A1
• Time in Patent Office: 1,835 Days
• Field of Search: None
• US Class Current: 707/770
• CPC Class Codes: G06F 2209/505 Clust
• CPC Class Codes: G06F 9/5061 Partitioning or combining of resources
• CPC Class Codes: G06N 5/046 Forward inferencing; Production systems

Abstract

A distributed computing system conforms to a multi-level, hierarchical organizational model. One or more control nodes provide for the efficient and automated allocation and management of computing functions and resources within the distributed computing system in accordance with the organization model. The model includes four distinct levels: fabric, domains, tiers and nodes that provide for the logical abstraction and containment of the physical components as well as system and service application software of the enterprise. A user, such as a system administrator, interacts with the control nodes to logically define the hierarchical organization of distributed computing system. The control node includes an automation subsystem having one or more rule engines that provide autonomic control of the application nodes in accordance with a set of one or more rules.

Claims

1. A distributed computing system comprising:

a plurality of application nodes interconnected via a communications network, wherein each application node comprises one or more programmable processors for executing software instructions;

a database storing data that defines a model for a hierarchical organization of the distributed computing system; and

a control node comprising;

one or more programmable processors;

a monitoring subsystem that collects status data from the application nodes,

wherein the status data represents an actual state for the application nodes;

a business logic tier that presents a user interface that allows an expected state of the application nodes to be defined; and

an automation subsystem comprising;

a sensor subsystem that receives the status data from the monitoring subsystem and aggregates the status data in accordance with the hierarchical organization defined by the model; and

one or more forward-chaining rule engines that provide autonomic control of the application nodes in accordance with a set of one or more rules and the hierarchical organization defined by the model,

wherein the forward-chaining rule engines receive expected state data representative of the expected state of the application nodes from the business logic tier, analyze the aggregated status data from the sensor subsystem, and apply the set of rules to produce action requests to the business logic tier to control the application nodes to reduce any difference between the actual state and the expected state,

wherein each of the one or more forward-chaining rule engines comprises a working memory that includes one or more local objects,

wherein each rule in the set of one or more rules includes a match condition and an implied action, wherein the set of one or more rules is compiled into a discrimination network that includes a translation of the match conditions of the set of one or more rules such that at least one redundant test is avoided during rule execution, and

wherein each of the one or more forward-chaining rule engines further comprises an execution engine that matches the set of one or more compiled rules within the discrimination network against the one or more local objects within the working memory.

2. The distributed computing system of claim 1, wherein the automation subsystem includes a rule compiler that compiles the rules into one or more discrimination networks.

3. The distributed computing system of claim 1, wherein the monitoring subsystem comprises a set of modular information collectors that collect the status data from the application nodes.

4. The distributed computing system of claim 3, wherein each of the information collectors conforms to a different communication protocol.

5. The distributed computing system of claim 3, wherein the application nodes are organized into tiers, and wherein each of the information collectors collects the status data from the application nodes of different tiers.

6. The distributed computing system of claim 3, wherein the automation unit registers collection requests with the monitoring subsystem, and the monitoring subsystem includes a monitoring engine that invokes the collectors in response to the collection requests.

7. The distributed computing system of claim 1, wherein the sensor subsystem communicates the aggregated data to the rule engines in accordance with a schedule.

8. The distributed computing system of claim 1, an effector subsystem that outputs action requests to the business logic tier in response to application of the rules by the rule engines.

9. The distributed computing system of claim 1, wherein the business logic tier includes a set of web services to control the application nodes.

10. The distributed computing system of claim 1, wherein the business logic tier presents an interface by which a user defines one or more goals for the expected state of the application nodes.

11. The distributed computing system of claim 10, wherein the business logic tier registers the goals with the automation unit for achievement by the rule engines in accordance with the rule set.

12. The distributed computing system of claim 10, wherein the automation unit automatically formulates dependencies between the goals.

13. The distributed computing system of claim 10,

wherein the business logic tier further comprises a task manager to invoke tasks in response to requests from the automation unit, and

wherein the business logic tier includes an event subsystem that tracks the progress of each of the tasks.

14. The distributed computing system of claim 10, wherein the goals include at least one of software image management goals, node allocation goals, goals for harvesting nodes, tier capacity goals, asset requirement goals and goals for gathering the status data.

15. The distributed computing system of claim 10, wherein the rules are units of code that conform with a rule language.

16. The distributed computing system of claim 1,

wherein the model specifies a fabric having a one or more domains, and wherein each domain has at least one tier that includes at least one of the application nodes, and

wherein the rule engines automatically produce action requests to configure the application nodes in accordance with a hierarchical organization defined by the model.

17. The distributed computing system of claim 16, wherein each tier represents a different computing function.

26. The distributed computing system of claim 1, wherein the execution engine collects all matching rules and matching local objects, selects a particular rule instantiation to fire, executes the selected instantiated rule, and propagates any changes to the working memory, and repeats the collection, selection, execution, and propagation processes until no more matching rule instantiations can be found.

18. A method comprising:

storing data that defines a model for a hierarchical organization of the distributed computing system;

receiving input that defines an expected state for a distributed computing system having a plurality of application nodes interconnected via a communications network;

receiving status data that represents an actual state for the distributed computing system;

processing rules with a set of forward-chaining rule engines to automatically determine operations for reducing any difference between the actual state and the expected state; and

applying the operations to the distributed computing system to control the application nodes in accordance with the rules and the hierarchical organization defined by the model.

19. The method of claim 18, wherein processing the rules with the set of forward-chaining rule engines comprises compiling the rules into one or more discrimination networks.

20. The method of claim 18, further comprising collecting the status data from the application nodes with a plurality of modular information collectors.

21. The method of claim 18, further comprising:

registering collection requests with a monitoring subsystem; and

executing a monitoring engine that invokes the collectors in response to the collection requests.

22. The method of claim 18, further comprising:

aggregating the state data in accordance with a hierarchical model of deployment of the application nodes; and

processing the aggregated state data with the rule engines.

23. The method of claim 18, further comprising:

presenting an interface by which a user defines the expected state as one or more goals; and

registering the goals with the rule engines.

24. The method of claim 23, wherein the model specifies a fabric having a one or more domains, and wherein each domain has at least one tier that includes at least one of the application nodes.

27. The method of claim 18, further comprising:

wherein each of the forward-chaining rule engines in the set of forward-chaining rule engines comprises a working memory that includes one or more local objects,

wherein each rule includes a match condition and an implied action, wherein the rules are compiled into a discrimination network that includes a translation of the match conditions of the rules such that at least one redundant test is avoided during rule execution, and

wherein processing the rules with the set of forward-chaining rule engines comprises matching the compiled rules within the discrimination network against the one or more local objects within the working memory.

28. The method of claim 18, wherein processing the rules with the set of forward-chaining rule engines further comprises:

collecting all matching rules and matching local objects,

selecting a particular rule instantiation to fire,

executing the selected instantiated rule,

propagating any changes to the working memory, and

repeating the collection, selection, execution, and propagation processes until no more matching rule instantiations can be found.

25. A computer-readable medium comprising instructions that cause a programmable processor to:

store data that defines a model for a hierarchical organization of the distributed computing system;

process rules with a set of forward-chaining rule engines to automatically determine operations for reducing differences between an actual state of a distributed computing system and an expected state; and

apply the operations to the distributed computing system to provide autonomic control of the application nodes in accordance with the rules and the hierarchical organization defined by the model.