Hardware/software based indirect time stamping methodology for proactive hardware/software event detection and control

• US 7,529,979 B2
• Filed: 12/12/2003
• Issued: 05/05/2009
• Est. Priority Date: 12/12/2003
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Dombrowa, Marc B., Sahoo, Ramendra K., Sugavanam, Krishnan, Hoenicke, Dirk I.
• Primary Examiner(s): Baderman; Scott T.
• Assistant Examiner(s): Truong; Loan
• Application Number:
  US10/735,412
• Publication Number:
  US 20050144532A1
• Time in Patent Office: 1,971 Days
• Field of Search: 714/39, 714/26, 714/37, 714/45
• US Class Current: 714/39
• CPC Class Codes: G06F 11/0751 Error or fault detection not based on redundancy power supply failures G06F1/30; network fault management H04L41/06
• CPC Class Codes: G06F 11/3452 Performance evaluation by statistical analysis
• CPC Class Codes: G06F 11/348 Circuit details, i.e. tracer hardware
• CPC Class Codes: G06F 11/3495 for systems
• CPC Class Codes: G06F 2201/835 Timestamp
• CPC Class Codes: G06F 2201/86 Event-based monitoring
• CPC Class Codes: G06F 2201/88 Monitoring involving counting

Abstract

An improved method and apparatus for time stamping events occurring on a large scale distributed network uses a local counter associated with each processor of the distributed network. Each counter resets at the same time globally so that all events are recorded with respect to a particular time. The counter is stopped when a critical event is detected. The events are masked or filtered in an online or offline fashion to eliminate non-critical events from triggering a collection by the system monitor or service/host processor. The masking can be done dynamically through the use of an event history logger. The central system may poll the remote processor periodically to receive the accurate counter value from the local counter and device control register. Remedial action can be taken when conditional probability calculations performed on the historical information indicate that a critical event is about to occur.

Claims

1. A method of analyzing events occurring on a distributed network comprising a plurality of processors, the method comprising:

during offline processing;

updating a recent history table with critical events and any associated non-critical events that may occur in the distributed network;

computing conditional probability values into a probability table, said probability values comprising joint probability values reflecting a probability that a sequence of two or more non-critical events happen before the critical event occurs;

periodically updating the probability table;

periodically examining the probability table to determine when online analysis of the non-critical event is possible; and

generating event masks for use in a masking mechanism to filter a subset of the non-critical events associated with the critical events so that online analysis can be carried out in real-time;

during online processing;

loading the conditional probability table and the event masks computed from offline analysis;

dynamically filtering the non-critical events using the masking mechanism comprising timeout and probability thresholds;

determining that the probability of the occurrence of the critical event has surpassed a threshold level using the conditional probability table;

migrating a process away from the critical event if it is determined that a timeout period has not elapsed; and

if it is determined that the timeout period has elapsed;

reloading the conditional probability tables; and

generating new event masks to filter another subset of the non-critical events.

2. The method of claim 1 wherein the step of dynamically filtering comprises:

using AND gates to filter and unfilter events based on online and offline analysis; and

using OR gates to record a final time stamp and event type in the recent history table.

3. The method of claim 1 further comprising an online processing step of:

producing a local counter value for each of the plurality of processors in the distributed network.

4. The method of claim 3 further comprising an online processing step of:

synchronizing the local counter value at each of the processors with a global clock.

5. The method of claim 4 further comprising an online processing step of:

freezing the local counter value for a processor when a critical event associated with the processor occurs.