SYSTEM AND METHOD FOR PROVIDING A SCALABLE SEMANTIC MECHANISM FOR POLICY-DRIVEN ASSESSMENT AND EFFECTIVE ACTION TAKING ON DYNAMICALLY CHANGING DATA

• US 20150286969A1
• Filed: 04/08/2015
• Published: 10/08/2015
• Est. Priority Date: 04/08/2014
• Status: Active Grant

• Current Assignee: Peraton, Inc.
• Original Assignee: Northrop Grumman Systems Corporation (Northrop Grumman Corporation)
• Inventors: Warner, Steven Alan, Kruba, Stephen Richard, Mohammad, Sohel, Karman, Alexander Francis, Kobes, Jason Christian
• Granted Patent: US 10,521,747 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06Q 10/0633 Workflow analysis

Abstract

A system for providing a scalable semantic mechanism for policy-driven assessment and action taking on dynamically changing data includes a policy context that is a collection of ontologies that capture policies of the enterprise in a semantic, declarative grammar, in a semantic database, a performance context that receives and stores sensor data output monitoring managed assets persisted in a time-series database, a workflow context that determines the workflow process necessary to manage the managed asset based on the governance policies and sensor data, implemented using a workflow engine that supports a declarative workflow language, a decision context that contains business rules encoded in a declarative grammar, implemented using a business rule engine and in which the business rules define conditions under which an asset is corrected, and an adjudication context that defines all associations between the four contexts that are necessary to monitor and manage the enterprise managed assets.

Claims

1. A system for providing a scalable semantic mechanism for policy-driven assessment and action taking on dynamically changing data, comprising:

a plurality of servers accessible via network connection by users of an enterprise, wherein the plurality of servers provide the execution environment for;

a policy context, wherein the policy context is a polymorphic collection of ontologies that capture governance policies of the enterprise in a semantic, declarative grammar and the policy context is persisted in a semantic database running on one or more of the plurality of servers;

a performance context, wherein the performance context receives and stores data output from sensors monitoring managed assets of the enterprise and the performance context is persisted in a time-series database running on one or more of the plurality of servers;

a workflow context, wherein the workflow context determines and directs the workflow process necessary to manage the managed asset based on the governance policies captured in the policy context and sensor data stored in the performance context and the workflow context is implemented using a workflow engine that supports a declarative workflow definition language and which is running on one or more of the plurality of servers;

a decision context, wherein the decision context contains business rules encoded in a declarative grammar and the decision context is implemented using a business rule engine running on one or more of the plurality of servers and which the business rules define conditions under which a managed asset or a managed asset setting must be started, stopped or assigned a specific value; and

an adjudication context, wherein the adjudication context defines all associations between the policy context, performance context, workflow context and decision context that are necessary to monitor and manage the enterprise managed assets, the adjudication context is semantic and declarative and the adjudication context is implemented using a semantic database running on one or more of the plurality of servers.

2. The system of claim 1 wherein the data output received from the sensors by the performance context includes events and metrics, wherein the events represent instantaneous transitions between discrete states of managed assets and metrics represent continuous transitions in the state of managed assets.

3. The system of claim 2 wherein performance context uses declarative and schema-less grammars to define events and metrics and the source managed assets for the received events and metrics.

4. The system of claim 1 wherein performance context time-stamps the data output received from the sensors and stores the data output in the time-series database.

5. The system of claim 1 further comprising a finite state machine that runs on one or more of the plurality of servers and which operates on and executes adjudication context, policy context, performance context, workflow context and decision context.

6. The system of claim 5 wherein the finite state machine is a self-assembling autonomic control system that is a stand-alone application that contain goals, preferences and algorithms to operate the system.

7. The system of claim 6 wherein self-assembling autonomic control system is declarative.

8. The system of claim 1 wherein system executes an observe, orient, decision and action (OODA) loop.

9. The system of claim 8 wherein the OODA loop includes performance context making direct interface invocations to sensors, adjudication context querying policy context, performance context, workflow context and decision context to ensure observations, decisions and actions are compliant with enterprise policies, performance context, workflow context and decision context querying one another to determine whether managed assets are compliant and whether to take corrective action and workflow context querying actuators that take corrective action with managed assets.

10. The system of claim 1 further comprising a plurality of actuators in communication with decision context and adjudication context, wherein the actuators take actions to bring managed assets into compliance with enterprise policies included in policy context.

11. The system of claim 1 further comprising a plurality of managed assets, wherein managed assets are anything that an enterprise has policies for monitoring and managing.

12. The system of claim 11 wherein the managed assets include servers, networks, firewalls, configuration items, databases, virtual machines, deployed applications and startup scripts.

13. The system of claim 1 wherein the plurality of servers are cloud computing servers that provide a cloud computing environment.

14. The system of claim 13 wherein the cloud computing environment provides the policy context, performance context, workflow context, decision context and adjudication context for the enterprise and a plurality of additional enterprises.

15. The system of claim 14 wherein the policy context, performance context, workflow context, decision context and adjudication context for the enterprise are separated from the policy context, performance contexts, workflow contexts, decision contexts and adjudications contexts for the plurality of additional enterprises using named graphs.

16. A method for providing a scalable semantic mechanism for policy-driven assessment and action taking on dynamically changing data, comprising:

defining a policy context for an enterprise, wherein the policy context is defined based on user inputs and policy document of an enterprise received by one or more servers that provide a semantic database in which the policy context is defined and the defining includes creating ontologies that capture governance policies of the enterprise in a semantic, declarative grammar;

defining a performance context for the enterprise, wherein defining the performance context for the enterprise is based on received user input, the policy context, an identification of managed assets of the enterprise, and the managed assets that must be managed per the policy context and the defining defines the performance context in a declarative grammar and deploying the performance context in a queryable time-series database that executes on the one or more servers;

defining a workflow context for the enterprise, wherein defining the workflow context is based on user input of actuators available to adjust setting on managed assets, known information regarding operation of such actuators, and the settings of managed assets and defining includes serializing the workflow context in a declarative grammar and deploying the workflow context in a queryable workflow engine that executes on the one or more servers;

defining a decision context for the enterprise, wherein defining a decision context for the enterprise includes setting thresholds and acceptable levels for managed assets and otherwise setting business rules for the enterprise and the defining includes serializing the decision context in a declarative grammar and deploying the decision context in a queryable business rules engine that executes on the one or more servers; and

defining an adjudication context for the enterprise, wherein the defining an adjudication context maps each managed asset to its parameters and links the policy context, performance context, workflow context and decision context and defining an adjudication context includes serializing the adjudication context in a declarative grammar and deploying the adjudication context in a queryable environment semantic database that executes on the one or more servers.

17. The method of claim 16 further comprising defining feedback control transfer functions for the enterprise executed by a finite state machine to effectuate an observe, orient, decide and act (OODA) loop, wherein the defining a feedback transfer functions includes serializing the functions in a declarative grammar and deploying the functions in a queryable finite state machine that executes on the one or more servers.

18. The method of claim 16 comprising receiving managed assets events and metrics from sensors and persisting the events and metrics in time-series database of performance context per enterprise policies set forth in policy context.

19. The method of claim 18 further comprising workflow context checking events and metrics and feeding the events and metrics values to the business rule engine of decision context.

20. The method of claim 19 further comprising:

adjudication context supplying parameters that the business rule engine needs to make decisions of whether action needs to be taken;

business rule engine of decision context deciding whether action needs to be taken for managed assets,

workflow context checking the adjudication context again to decide which corrective action workflow to apply;

adjudication context populating the parameters that the corrective workflow instance needs to instruct corrective workflow; and

workflow context invoking one or more actuators to effectuate corrective action on managed assets.