General event stamping scheme

• US 6,332,117 B1
• Filed: 10/22/1998
• Issued: 12/18/2001
• Est. Priority Date: 10/22/1998
• Status: Expired due to Term

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Urquhart, Robert John, Berry, Robert Francis, Levine, Frank Eliot, Hussain, Riaz Y., John, Chester Charles Jr.
• Primary Examiner(s): Hoff, Marc S.
• Assistant Examiner(s): Barbee, Manuel L.
• Application Number:
  US09/177,035
• Time in Patent Office: 1,153 Days
• Field of Search: 702/81, 702/119, 702/122, 702/123, 702/176, 702/177-180, 702/182-188, 702/FOR.135, 702/139, 714/25, 714/33, 714/39, 714/45, 714/47, 717/6
• US Class Current: 702/187
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 11/3466 Performance evaluation by tracing or monitoring
• 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
• CPC Class Codes: G06F 2201/885 Monitoring specific for caches

Abstract

During the recording of events occurring within the operation of a data processing system, a user-selected metric of interest is recorded along with the trace data for each recorded event. The metric of interest is a monotonically increasing variable that provides a non-time relationship between the recorded events. This relationship may be time or non-time based.

Claims

1. In a data processing system, a method comprising the steps of:

collecting events from a monitored process over a span of time;

stamping each of the events with a user-selectable metric of interest; and

storing each of the events with its respective metric of interest.

2. The method as recited in claim 1, wherein the user-selectable metric of interest is monotonically increasing over a span of time.

3. The method as recited in claim 2, wherein the metric of interest is not a timestamp.

4. The method as recited in claim 2, wherein the events are trace events that are stored in a trace buffer with the stamped metric of interest.

5. The method as recited in claim 4, further comprising the step of postprocessing the trace events.

6. The method as recited in claim 4, wherein the metric of interest is selected from a group of time and non-time based metrics of interest.

7. A data processing system comprising:

a processor coupled to a storage device, a memory device, an input device, and an output device via a bus system, wherein the processor is operable for running a specified process and an event recording program, the processor further comprising;

circuitry for monitoring the process;

circuitry for collecting events from the monitored process over a span of time;

circuitry for stamping each of the events with a non-time, increasing metric of interest; and

circuitry for storing each of the events with its respective metric of interest in the memory device.

8. The system as recited in claim 7, wherein the non-time, increasing metric of interest is monotonically increasing over the span of time.

9. The system as recited in claim 7, wherein the metric of interest is not a timestamp.

10. The system as recited in claim 7, wherein the events are trace events that are stored in a trace buffer with the stamped metric of interest.

11. The system as recited in claim 7, wherein the metric of interest is supplied by a counter implemented within the processor.

12. A computer program product adaptable for storage in a computer readable medium and operable for recording events occurring within a process operating in a data processing system, the computer program product comprising:

programming operable for collecting events from a monitored process over a span of time;

programming operable for stamping each of the events with a user-selectable metric of interest; and

programming operable for storing each of the events with its respective metric of interest.

13. The computer program product as recited in claim 12, wherein the user-selectable metric of interest is monotonically increasing over the span of time.

14. The computer program product as recited in claim 13, wherein the metric of interest is not a timestamp.

15. The computer program product as recited in claim 13, wherein the events are trace events that are stored in a trace buffer with the stamped metric of interest.

16. The computer program product as recited in claim 13, wherein the metric of interest is supplied by a counter implemented within the data processing system.

17. The computer program product as recited in claim 13, further comprising:

programming operable for initializing a trace program in the data processing system, including defining a trace buffer operable for storing the events; and

programming operable for specifying the metric of interest.

18. The computer program as recited in claim 12, wherein the metric of interest may be selected from a group consisting of time and non-time based metrics of interest.

19. In a data processing system, a method for recording an event comprising the steps of:

reading configuration parameters to determine the size and location of a trace buffer;

initializing said trace buffer and a system trace program to trace user specified events;

collecting said user specified events from said trace program;

stamping said user specified events with a user specified metric of interest; and

storing said user specified events with said user specified metric of interest.

20. The method as recited in claim 19, wherein said user specified metric of interest is not a timestamp.

21. The method as recited in claim 19, wherein said user specified metric of interest is monotonically increasing over a span of time.

22. A computer program product adaptable for storage in a computer readable medium and operable for recording events occurring within a process operating in a data processing system, the computer program product comprising:

programming operable for reading configuration parameters to determine the size and location of a trace buffer;

programming operable for initializing said trace buffer and a system trace program to trace user specified events;

programming operable for collecting said user specified events from said trace program;

programming operable for stamping said user specified events with a user specified metric of interest; and

programming operable for storing said user specified events with said user specified metric of interest.

23. The computer program product as recited in claim 22, wherein said user specified metric of interest is not a timestamp.

24. The computer program product as recited in claim 22, wherein said user specified metric of interest is monotonically increasing over a span of time.