System and method for problem determination using dependency graphs and run-time behavior models

• US 7,437,611 B2
• Filed: 10/21/2004
• Issued: 10/14/2008
• Est. Priority Date: 10/21/2004
• Status: Expired due to Fees

• Current Assignee: GlobalFoundries, Inc.
• Original Assignee: International Business Machines Corporation
• Inventors: Gupta, Manish, Neogi, Anindya, Sailer, Anca, Appleby, Karen, Agarwal, Manoj K., Kar, Gautam
• Primary Examiner(s): Chu; Gabriel L
• Application Number:
  US10/968,917
• Publication Number:
  US 20060101308A1
• Time in Patent Office: 1,454 Days
• Field of Search: None
• US Class Current: 714/26
• CPC Class Codes: G06F 11/0709 in a distributed system consisting of a plurality of standalone computer nodes, e.g. clusters, client-server systems
• CPC Class Codes: G06F 11/079 Root cause analysis, i.e. error or fault diagnosis in a hardware test environment G06F11/22; in a software test environment G06F11/36
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/87 Monitoring of transactions

Abstract

A problem determination system and method reduces the time and effort required by system administrators to trouble shot transaction processing difficulties in a large distributed I/T environment by monitoring system component performance and computing operational performance threshold limits based on dependency graphs for the system. Using this data, a prioritized list of suspect components for the I/T system is generated.

Claims

1. A computer implemented method for performing problem determination within a distributed transaction system comprising the steps of:

computing operational limits for a set of resources in a distributed transaction system;

determining a poor performance status for said resources relative to said operational limits;

generating a list of root cause components which are likely to be causing said poor performance status;

prioritizing said list of root cause components; and

providing said prioritized list of root cause components to a system administrator,

wherein the step of computing operational limits for a set of resources in a distributed transaction system further comprises the steps of;

capturing monitored information from said distributed transaction system;

computing individual component level thresholds from user level SLAs; and

generating dynamic dependency data and said monitored information.

2. A computer implemented method for performing problem determination within a distributed transaction system comprising the steps of:

computing operational limits for a set of resources in a distributed transaction system;

determining a poor performance status for said resources relative to said operational limits;

generating a list of root cause components which are likely to be causing said poor performance status;

prioritizing said list of root cause components; and

providing said prioritized list of root cause components to a system administrator,

wherein the step of determining a poor performance status for said resources relative to said operational limits further comprises the steps of;

obtaining monitoring information relative to performance of a transaction;

identifying a sub-set of resources that are used to perform said transaction from said set of resources in said distributed transaction system;

obtaining performance information for each resource of said sub-set of resources; and

comparing said monitoring information with said performance information to make performance status determination.

3. A computer implemented method for performing problem determination within a distributed transaction system comprising the steps of:

computing operational limits for a set of resources in a distributed transaction system;

determining a poor performance status for said resources relative to said operational limits;

generating a list of root cause components which are likely to be causing said poor performance status;

prioritizing said list of root cause components; and

providing said prioritized list of root cause components to a system administrator,

wherein the step of generating a list of root cause components which are likely to be causing said poor performance status further comprises the steps of;

selecting one of several algorithms for calculating a severity value for each of said components; and

creating a priority list of said set root cause components based on severity values.