Active probing for real-time diagnosis

• US 7,389,347 B2
• Filed: 04/16/2004
• Issued: 06/17/2008
• Est. Priority Date: 04/16/2004
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Rish, Irina, Odintsova, Natalia V., Grabarnik, Genady, Ma, Sheng, Brodie, Mark A.
• Primary Examiner(s): Patel; Haresh N
• Application Number:
  US10/826,575
• Publication Number:
  US 20050235057A1
• Time in Patent Office: 1,523 Days
• Field of Search: 709200-203, 709217-228, 719/310, 702/168
• US Class Current: 709/224
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 43/12 Network monitoring probes

Abstract

Improved problem diagnosis techniques for use in accordance with computing systems, e.g., distributed computing systems, are disclosed. In one aspect of the invention, a technique for diagnosing a problem associated with a computing system comprises the following steps/operations. One or more probes are executed in accordance with at least a portion of a previously selected probe schedule. When a result of one or more of the probes of the previously selected probe schedule indicates, at least, a potential problem associated with the computing system, one or more probes which optimize at least one criterion are selected in real-time. The one or more selected probes are executed so as to diagnose the potential problem.

Claims

1. A computer-implemented method for diagnosing a problem associated with a computing system, the method comprising the steps of:

executing one or more probes in accordance with at least a portion of a previously selected probe schedule for active probing;

when a result of one or more of the probes of the previously selected probe schedule indicates, at least, a potential problem associated with the computing system, selecting in real-time one or more probes which optimize at least one criterion, wherein the step of selecting in real-time one or more probes which optimize at least one criterion further comprises the step of selecting in real-time one or more probes which maximize information gain relating to the potential problem;

executing the one or more selected probes so as to diagnose the potential problem;

analyzing results of the execution of the one or more selected probes using a probabilistic inference and using one or more prior fault probabilities for one or more system components;

wherein prior to the execution of the probes, a problem detection probe set (DPS) and a problem localization probe set (LPS) are pre selected, wherein probes of the DPS are intended to cover any given problem and probes of the LPS are intended to localize a problem detected by a probe of the DPS; and

repeating the step of selecting in real-time one or more probes which optimize at least one criterion and the step of analyzing results of the execution of the one or more selected probes until a particular level of diagnostic confidence is reached.