Dismiss
InnovationQ will be updated on Sunday, Oct. 22, from 10am ET - noon. You may experience brief service interruptions during that time.
Browse Prior Art Database

Adaptive Algorithm for Automating Problem Determination.

IP.com Disclosure Number: IPCOM000021316D
Original Publication Date: 2004-Jan-13
Included in the Prior Art Database: 2004-Jan-13
Document File: 3 page(s) / 88K

Publishing Venue

IBM

Abstract

Existing debugging tools can generate data that are extensive in size and complex in content. Analyzing/validating such data could be time consuming and inefficient. Disclosed is an implementation of an algorithm that automates the validation process. The debugging tools are still required for capturing the data, however, once a set of rules is established, validation against these rules using the proposed algorithm is much more efficient and considerably less time consuming.

This text was extracted from a PDF file.
At least one non-text object (such as an image or picture) has been suppressed.
This is the abbreviated version, containing approximately 55% of the total text.

Page 1 of 3

Adaptive Algorithm for Automating Problem Determination.

    Debugging PC related issues could be an elusive and time consuming task. The need to devise an algorithm to expedite the debugging process for these issues has become of a particular importance. Although there are several industrial tools to capture device activities, such as Logic analyzers, Oscilloscopes, Serial ATA analyzer, etc., the captured traces (data) from such tools could be enormous in size and tedious to analyze. Engineers are spending numerous hours trying to decipher these captures and root-cause the issue on hand. Once the root cause is found a rule could be established. The proposed technique is an implementation of an algorithm that takes the capture generated by the debugging tool, then by choosing the problem device to analyze, the captured data could be validated against the set of rules for that device using a "rules checker program". If any of the rules for that device was violated a flag is generated with a corrective action presented to the user. On the other hand, if there were no violations and further analysis of the capture helps with the root cause, a new rule for that device is established and added to the rule-pool to check against. The following is the state machine and flow chart describing the adaptive process mentioned above:
State Machine:

1) Start with a most likely device tree
2)Are there any devices to check?

2a) If yes, continue to 3

2b) If NO, manual intervention required. Add a rule into the rules vector.
3) Push the logic to capture the device bus activities
4) Input the device bus activities into the rules checker program
5) Did the rules checker find any violation

5a) If Yes did the violation contribute to the root cause of the problem

5b) If Yes, root cause found Exit

5c) If No, flag an error and continue

5d) If No, go the next device in the tree and go to 2)

Page 2 of 3

Start with a most likely device

tree.

Manually debug problem and add to rules vector.

Any devices to check?

Did the rules checker find any violation?

Did the violation contribute to root cause?

Push the logic to capture the device bus acti...