Methods and apparatus for detection of performance conditions in processing system

• US 7,908,546 B2
• Filed: 05/11/2007
• Issued: 03/15/2011
• Est. Priority Date: 05/11/2007
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Alcatel-Lucent USA, Inc. (Nokia Corporation)
• Inventors: Runyon, James Philip, Rauscher, Karl F., Krock, Richard E., Goldman, Stuart Owen
• Primary Examiner(s): Chaudry, M. Mujtaba K
• Assistant Examiner(s): Ahmed, Enam
• Application Number:
  US11/747,526
• Publication Number:
  US 20080282134A1
• Time in Patent Office: 1,404 Days
• Field of Search: 714/798, 714/47, 700/32, 370/241
• US Class Current: 714/798
• CPC Class Codes: G06F 11/3409 for performance assessment
• CPC Class Codes: G06F 2201/81 Threshold
• CPC Class Codes: G06F 2201/87 Monitoring of transactions
• CPC Class Codes: G06F 2201/88 Monitoring involving counting
• CPC Class Codes: H04L 43/10 Active monitoring, e.g. heartbeat, ping or trace-route
• CPC Class Codes: H04L 43/16 Threshold monitoring

Abstract

Techniques are disclosed for detection of performance conditions in processing systems. For example, a method of detecting a performance condition in at least one particular processing device of a processing system having a plurality of processing devices includes the following steps. Data is input to a data structure associated with the particular processing device, over a given time period. The input data may be a buffer or a bucket. The input data represents data associated with the execution of at least one function performed by the particular processing device. The given time period includes the time period between consecutive heartbeat signals transmitted by the particular processing device. At least a portion of the input data is removed from the data structure associated with the particular processing device, near the end of the given time period. The removed input data is compared to an expected function execution level. An alarm signal is generated, when warranted, based on the comparison of the removed input data to the expected function execution level such that a performance condition in the particular processing device is determinable.

Claims

1. A method of detecting a performance condition in at least one particular processing device of a processing system, wherein the processing system comprises a plurality of processing devices, the method comprising the steps of:

inputting data to a data structure associated with the particular processing device, over a given time period, the input data representing data associated with the execution of at least one function performed by the particular processing device, and the given time period comprising the time period between consecutive heartbeat signals transmitted by the particular processing device;

removing at least a portion of the input data from the data structure associated with the particular processing device, near the end of the given time period;

comparing the removed input data to an expected function execution level; and

generating an alarm signal, when warranted, based on the comparison of the removed input data to the expected function execution level such that a performance condition in the particular processing device is determinable.

2. The method of claim 1, wherein the input data represents a count of the number of times the particular processing device completes execution of the function.

3. The method of claim 2, wherein the portion of input data removed from the data structure, for the given time period, is determined as a percentage of the number of counts input to the data structure for the given time period.

4. The method of claim 1, wherein the amount of input data removed from the data structure is adjustable for a subsequent time period.

5. The method of claim 4, wherein adjustment of the amount of input data removed from the data structure for the subsequent time period is a function of a previous time period.

6. The method of claim 1, wherein the amount of input data removed from the data structure comprises a minimum value greater than zero.

7. The method of claim 1, wherein the generated alarm signal is representative of an initial warning indicative of an initial detection of a potential performance problem in the particular processing device.

8. The method of claim 7, wherein the occurrence of multiple initial warnings over multiple given time periods results in generation of at least a second alarm signal that is representative of a warning of higher priority than the initial warning.

9. The method of claim 1, wherein the data structure is maintained internal to the particular processing device.

10. The method of claim 1, wherein the data structure is maintained external to the particular processing device.

11. The method of claim 1, further comprising the step of transmitting the alarm signal along with the heartbeat signal to at least one of a system administrator and a central monitoring system.

12. The method of claim 1, wherein a size of the data structure is adjustable for a subsequent time period.

13. The method of claim 1, wherein each of the plurality of processing devices in the processing system have respective data structures associated therewith such that the inputting, removing, comparing, and generating steps are performable for each of the plurality of processing devices.

14. The method of claim 1, wherein the processing system is at least part of a communications system or a computing system.

15. Apparatus for detecting a performance condition in at least one particular processing device of a processing system, wherein the processing system comprises a plurality of processing devices, the apparatus comprising:

a memory; and

at least one processor coupled to the memory and operative to;

(i) input data to a data structure associated with the particular processing device, over a given time period, the input data representing data associated with the execution of at least one function performed by the particular processing device, and the given time period comprising the time period between consecutive heartbeat signals transmitted by the particular processing device;

(ii) remove at least a portion of the input data from the data structure associated with the particular processing device, near the end of the given time period;

(iii) compare the removed input data to an expected function execution level; and

(iv) generate an alarm signal, when warranted, based on the comparison of the removed input data to the expected function execution level such that a performance condition in the particular processing device is determinable.

16. The apparatus of claim 15, wherein the input data represents a count of the number of times the particular processing device completes execution of the function.

17. The apparatus of claim 15, wherein the amount of input data removed from the data structure is adjustable for a subsequent time period.

18. The apparatus of claim 15, wherein the at least one processor is further operative to transmit the alarm signal along with the heartbeat signal to at least one of a system administrator and a central monitoring system.

19. The apparatus of claim 15, wherein a size of the data structure is adjustable for a subsequent time period.

20. A processing system, comprising:

a plurality of processing devices;

wherein a performance condition in at least one particular processing device of the plurality of processing devices is detectable by at least one processor configured to;

(i) input data to a data structure associated with the particular processing device, over a given time period, the input data representing data associated with the execution of at least one function performed by the particular processing device, and the given time period comprising the time period between consecutive heartbeat signals transmitted by the particular processing device;

(ii) remove at least a portion of the input data from the data structure associated with the particular processing device, near the end of the given time period;

(iii) compare the removed input data to an expected function execution level; and

(iv) generate an alarm signal, when warranted, based on the comparison of the removed input data to the expected function execution level such that a performance condition in the particular processing device is determinable.