KnowOps: Towards an Embedded Knowledge ... - Semantic Scholar
KnowOps: Towards an Embedded Knowledge Base for Network Management and Operations Xu Chen*, Yun Mao*, Z. Morley Mao+, Kobus Van der Merwe* *AT&T
Labs – Research
+The
© 2007 AT&T Knowledge Ventures. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Knowledge Ventures.
University of Michigan – Ann Arbor
What is Network Management? • In short – Keep the network in a “healthy” state – Deliver SLA-compliant services for customers
• Not so short – Planned maintenance/upgrade – Fault management – Configuration management – Traffic/performance management – Security management – …
Page 2
© 2007 AT&T Knowledge Ventures. All rights reserved.
Simplified View of Network Management Systems Event Manager Ticketing System Network Inventory
Process Automation
Configuration Management
Event Correlation
Network Instrumentation
Network Interface
Network
Page 3
© 2007 AT&T Knowledge Ventures. All rights reserved.
Simplified View of Network Management Systems Event Knowledge Base Manager
Ticketing System
Vendor configuration example Network - Captured in text-based documents Inventory
Process Automation
Event Correlation
Network Instrumentation
Provider service design documents Configuration
Management - Difficult to keep in-sync across systems
OperationalNetwork Interface Experience/Domain Knowledge
Network
Page 4
- Require manual work to inject from documentation to management systems
© 2007 AT&T Knowledge Ventures. All rights reserved.
KnowOps: Using Shared and Machinereadable Knowledge Base Event Manager Ticketing System Network Inventory
Process Automation
Event Correlation
Network Instrumentation
Embedded Knowledge Base
Network
Page 5
© 2007 AT&T Knowledge Ventures. All rights reserved.
Configuration Management
Network Interface
COOLAID [CoNEXT 2010] COOLAID Views
Vendors database
Rules
Service Provider
Network
• Capture domain knowledge in a declarative language – Vendors: protocol mechanisms, dependencies – Service providers: service realizations, misconfigurations
• Automated reasoning mechanisms decoupled from the rules – Bottom-up reasoning – Top-down reasoning Page 6
© 2007 AT&T Knowledge Ventures. All rights reserved.
Shared Knowledge Base in KnowOps
- What to do? - What should be avoided?
- What events should be correlated? - What time windows should be used?
- What to monitor? - What to alarm?
Page 7
Service Requirements Process Automation
Vendor Rules Event Correlation
Network Instrumentation
Service Provider Rules
Network Configuration
Network © 2007 AT&T Knowledge Ventures. All rights reserved.
Preliminary Experience • DROOLS: open-source business logic – Rule engine – Process automation – Event correlation – Optimization/Planning
Page 8
© 2007 AT&T Knowledge Ventures. All rights reserved.
Example VPLS
MPLS, RSVP, OSPF
P PE
P
BGP Session
PE
LSP Connection VPLS Connection Site1
Page 9
Site2
© 2007 AT&T Knowledge Ventures. All rights reserved.
Example: VPLS VPLS
LSP
BGP
MPLS
RSVP
Interface Configuration
Page 10
© 2007 AT&T Knowledge Ventures. All rights reserved.
OSPF
Rules for OSPF Dependency on interface configuration
Rule 1
Rule 2
Page 11
© 2007 AT&T Knowledge Ventures. All rights reserved.
Planned Maintenance Example Automation Process Pre-maintenance check Device Vendor Rules
Disable related alarms
Service Provider Rules
“Disrupting VPLS service should raise a warning.”
Shut down interface
Page 12
© 2007 AT&T Knowledge Ventures. All rights reserved.
Conclusions • Take-aways – Knowledge transfer in current management systems are mostly text-based, thus costly and error-prone to build and maintain – We should build management systems based on a machinereadable, shared, and embedded knowledge base
• Challenges – What does the knowledge base really look like – Better integrate different contributors – Migrate from existing systems
• Future (on-going) work – Drools-based implementation – Application to mobility management tasks
Page 13
© 2007 AT&T Knowledge Ventures. All rights reserved.
Questions? Comments? • Thanks!
Page 14
© 2007 AT&T Knowledge Ventures. All rights reserved.
Labs – Research
+The
© 2007 AT&T Knowledge Ventures. All rights reserved. AT&T and the AT&T logo are trademarks of AT&T Knowledge Ventures.
University of Michigan – Ann Arbor
What is Network Management? • In short – Keep the network in a “healthy” state – Deliver SLA-compliant services for customers
• Not so short – Planned maintenance/upgrade – Fault management – Configuration management – Traffic/performance management – Security management – …
Page 2
© 2007 AT&T Knowledge Ventures. All rights reserved.
Simplified View of Network Management Systems Event Manager Ticketing System Network Inventory
Process Automation
Configuration Management
Event Correlation
Network Instrumentation
Network Interface
Network
Page 3
© 2007 AT&T Knowledge Ventures. All rights reserved.
Simplified View of Network Management Systems Event Knowledge Base Manager
Ticketing System
Vendor configuration example Network - Captured in text-based documents Inventory
Process Automation
Event Correlation
Network Instrumentation
Provider service design documents Configuration
Management - Difficult to keep in-sync across systems
OperationalNetwork Interface Experience/Domain Knowledge
Network
Page 4
- Require manual work to inject from documentation to management systems
© 2007 AT&T Knowledge Ventures. All rights reserved.
KnowOps: Using Shared and Machinereadable Knowledge Base Event Manager Ticketing System Network Inventory
Process Automation
Event Correlation
Network Instrumentation
Embedded Knowledge Base
Network
Page 5
© 2007 AT&T Knowledge Ventures. All rights reserved.
Configuration Management
Network Interface
COOLAID [CoNEXT 2010] COOLAID Views
Vendors database
Rules
Service Provider
Network
• Capture domain knowledge in a declarative language – Vendors: protocol mechanisms, dependencies – Service providers: service realizations, misconfigurations
• Automated reasoning mechanisms decoupled from the rules – Bottom-up reasoning – Top-down reasoning Page 6
© 2007 AT&T Knowledge Ventures. All rights reserved.
Shared Knowledge Base in KnowOps
- What to do? - What should be avoided?
- What events should be correlated? - What time windows should be used?
- What to monitor? - What to alarm?
Page 7
Service Requirements Process Automation
Vendor Rules Event Correlation
Network Instrumentation
Service Provider Rules
Network Configuration
Network © 2007 AT&T Knowledge Ventures. All rights reserved.
Preliminary Experience • DROOLS: open-source business logic – Rule engine – Process automation – Event correlation – Optimization/Planning
Page 8
© 2007 AT&T Knowledge Ventures. All rights reserved.
Example VPLS
MPLS, RSVP, OSPF
P PE
P
BGP Session
PE
LSP Connection VPLS Connection Site1
Page 9
Site2
© 2007 AT&T Knowledge Ventures. All rights reserved.
Example: VPLS VPLS
LSP
BGP
MPLS
RSVP
Interface Configuration
Page 10
© 2007 AT&T Knowledge Ventures. All rights reserved.
OSPF
Rules for OSPF Dependency on interface configuration
Rule 1
Rule 2
Page 11
© 2007 AT&T Knowledge Ventures. All rights reserved.
Planned Maintenance Example Automation Process Pre-maintenance check Device Vendor Rules
Disable related alarms
Service Provider Rules
“Disrupting VPLS service should raise a warning.”
Shut down interface
Page 12
© 2007 AT&T Knowledge Ventures. All rights reserved.
Conclusions • Take-aways – Knowledge transfer in current management systems are mostly text-based, thus costly and error-prone to build and maintain – We should build management systems based on a machinereadable, shared, and embedded knowledge base
• Challenges – What does the knowledge base really look like – Better integrate different contributors – Migrate from existing systems
• Future (on-going) work – Drools-based implementation – Application to mobility management tasks
Page 13
© 2007 AT&T Knowledge Ventures. All rights reserved.
Questions? Comments? • Thanks!
Page 14
© 2007 AT&T Knowledge Ventures. All rights reserved.
