System and method for problem resolution in communications networks

• US 7,475,130 B2
• Filed: 12/23/2004
• Issued: 01/06/2009
• Est. Priority Date: 12/23/2004
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Silverman, Robert M.
• Primary Examiner(s): Nguyen; Phuoc H
• Application Number:
  US11/019,454
• Publication Number:
  US 20060143496A1
• Time in Patent Office: 1,475 Days
• Field of Search: 709/217, 709/223, 709/224, 709/235
• US Class Current: 709/223
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: H04L 41/0663 Performing the actions predefined by failover planning, e.g. switching to standby network elements
• CPC Class Codes: H04L 41/16 using machine learning or artificial intelligence
• CPC Class Codes: H04L 43/08 Monitoring or testing based on specific metrics, e.g. QoS, energy consumption or environmental parameters
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/0876 Network utilisation, e.g. volume of load or congestion level
• CPC Class Codes: H04L 43/16 Threshold monitoring
• CPC Class Codes: H04Q 3/0083 Network planning or design; Modelling of planned or existing networks

Abstract

A system and method for using network asset management data including device type, location, and link speed for modeling network performance and identifying performance issues. Functionally complete mathematical characterization of a system may be employed to create a system for problem recognition and resolution, wherein disruptive changes to physical and logical system components are only performed when necessary. Mathematical analysis of pings, MIB, and other asset data may be performed separately and the results compared to “self-check”, form, and verify the complete mathematical characterization of a network system. The invention may provide projections of performance impacts based upon proposed changes to components in the network. Performance metrics, ping times, MIB data, link speeds, propagation delays, device latency, serialization rates, conversational speeds, conversational utilization and other data are used to model a network for determining or isolating problematic issues and to forecast performance impacts of potential changes to the network.

Claims

1. A method of managing a network, comprising the steps of:

modeling performance of a network having one or more network components to locate an instance of a mathematically impaired network component based on any combination of network speed, latency, tuning and utilization;

providing predictive modeling of potential changes to the network, including effects on performance of additional traffic, link speed increases, and chances in distance between clients and servers, thereby increasing a speed of problem resolution of one or more network components that slow overall performance across network connections due to queue build up, propagation delay, or slow serialization or processing of network traffic; and

modifying one or more parameters associated with the one or more network components based on the modeling to improve network performance.

2. The method of claim 1, further comprising the step of employing trial and error changes to remedy a network problem, when the modeling fails to locate the instance of the mathematically impaired network component.

3. The method of claim 1, wherein the mathematically impaired component comprises a measured or computed deficiency associated with the one or more network components.

4. The method of claim 1, further comprising reconfiguring the one or more components to remedy the network problem based on the modeling.

5. The method of claim 1, further comprising projecting what effect changes to the one or more network components on a link in the network has on performance of an end-to-end path.

6. The method of claim 1, wherein the modeling performance includes baseline performance and current performance levels of the network.

7. The method of claim 1, wherein the predictive modeling is what-if modeling.

8. A method for managing a network performance, comprising the steps of:

creating a baseline network model of a network having one or more network components;

adding MIB data to the baseline network model to create a model of current conditions;

pinging a path in the network to verify the model of current conditions;

identifying optimal tuning for the path based on the model of current conditions verified by the pinging;

modifying one or more parameters for at least one of the one or more network components based on the identified optimal tuning to improve network performance at least along the path; and

creating a predictive model of potential changes to the network, including effects on performance of additional traffic, link speed increases, and chances in distance between clients and servers, thereby increasing a speed of problem resolution of one or more network components that slow overall performance across network connections due to queue build up, propagation delay, or slow serialization or processing of network traffic.

9. The method of claim 8, further comprising the steps of:

identifying hops in the network with excessive latency; and

identifying network speed and queuing chokepoints in the path.

10. The method of claim 8, further comprising determining the network path in use for suspected instances of current poor performance based on at least any one of user documentation, detected activity and measured performance.

11. The method of claim 8, further comprising the steps of:

remediating any mathematical problem including at least any one of a computed problem and a measured problem; and

verifying whether a change based on the remediating improves the network performance.

12. The method of claim 8, further comprising checking whether one of the one or more network components is broken or unreachable.

13. A system for managing networks, comprising:

means for modeling performance of a network having one or more network components to locate an instance of a mathematically impaired network component based at least on any one of network speed, latency, tuning and utilization;

means for providing predictive modeling of potential changes to the network, including effects on performance of additional traffic, link speed increases, and changes in distance between clients and servers, thereby increasing a speed of problem resolution of one or more network components that slow overall performance across network connections due to queue build up, propagation delay, or slow serialization or processing of network traffic; and

means for modifying one or more network components including at least any one of a network component parameter and network component configuration based on the modeling to improve network performance at least along a path of the network.

14. The system of claim 13, further comprising means for employing trial and error changes to remedy a network problem when the means for modeling fails to locate the instance of the mathematically impaired network component.

15. The system of claim 13, wherein the mathematically impaired component comprises a measured or computed deficiency associated with the one or more network components.

16. The system of claim 13, further comprising means for reconfiguring the one or more components to remedy the network problem based on the modeling.

17. The system of claim 13, further comprising means for measuring and verifying a suspected problem in the network.

18. The system of claim 13, further comprising means for identifying hops and excessive latency in the network.

19. The system of claim 13, further comprising:

means for identifying optimal tuning for at least along the path; and

means for identifying speed and queuing chokepoints in the network.

20. The system of claim 13, further comprising:

means for creating a baseline model; and

means for creating a current condition model with management information base (MIB) data.

21. The system of claim 13, further comprising means for modeling network performance to isolate a performance problem.

22. A computer program product comprising a computer usable medium having readable program code tangibly embodied in the medium and executable by a computer, the computer program product includes at least one component to:

model performance of a network having one or more network components to locate an instance of a mathematically impaired network component based at least on any one of network speed, latency, tuning and utilization;

create a predictive model of potential changes to the network, including effects on performance of additional traffic, link speed increases, and changes in distance between clients and servers, thereby increasing a speed of problem resolution of one or more network components that slow overall performance across network connections due to queue build up, propagation delay, or slow serialization or processing of network traffic; and

modify one or more parameters associated with the one or more network components based on the modeling to improve network performance.