SYSTEM AND METHOD FOR PREDICTIVE NETWORK MONITORING

• US 20120023041A1
• Filed: 08/06/2008
• Published: 01/26/2012
• Est. Priority Date: 08/08/2007
• Status: Abandoned Application

• Current Assignee: Mark Zlochin, Ofer Shemesh, Yoram Kariv
• Original Assignee: Mark Zlochin, Ofer Shemesh, Yoram Kariv
• Inventors: Kariv, Yoram, Zlochin, Mark, Shemesh, Ofer
• Application Number:
  US12/672,520
• Publication Number:
  US 20120023041A1
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3447 Performance evaluation by modeling
• CPC Class Codes: G06F 11/3457 Performance evaluation by simulation
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: H04L 41/142 using statistical or mathematical methods
• CPC Class Codes: H04L 41/145 involving simulating, designing, planning or modelling of a network
• CPC Class Codes: H04L 41/147 for predicting network behaviour

Abstract

A system and a method for at least predicting a trend toward a reduction in performance of a computer and/or a computer network. Preferably, the system and method is able to predict a trend toward a potential failure of a computer and/or a computer network.

Claims

1. A method for at least predicting a trend toward reduced performance of a computer network, comprising:

Modeling behavior of the computer network; and

Predicting the trend according to said modeled behavior.

2. The method of claim 1, wherein said modeling comprising:

Determining a plurality of potential models for said behavior; and

Predicting the trend from said plurality of potential models.

3. The method of claim 2, wherein said predicting further comprises:

Combining predictions from said plurality models to form a combination to predict the trend.

4. The method of claim 3 wherein said combination is composed of weighted models.

5. The method of claim 4 wherein said weighted model affects said combination result according to its weight.

7. The method of claim 1, wherein said behavior of the computer network comprises behavior of at least one computer on the computer network.

8. The method of claim 7, wherein said behavior of the computer network comprises behavior of a plurality of computers interacting through the computer network.

9. The method claim 8, further comprising:

Monitoring the computer network to obtain data regarding one or more functions of the network; and

Improving a predictive model according to said data.

10. The method of claim 9, wherein said monitoring further comprises cleaning the data to filter out the noise in the data.

11. The method of claim 10, wherein said cleaning the data further comprises constructing a complex Bayesian distribution;

and reducing noise in the data according to correlation between a plurality of variables according to said distribution.

12. The method of claim 11, wherein said improving said predictive model comprises subjecting said predictive model to a learning period for processing historical data;

and examining performance of said predictive model with real time data.

13. The method of claim 11, further comprising:

issuing an alert if a predicted performance function of the network falls below or above a threshold.

6. A method for improving accuracy of predicting a trend toward reduced performance of a computer network, comprising:

Weighting each model according to accuracy of prediction;

Removing at least one potential model according to at least one criterion; and

Adding at least one potential new model.