Method and apparatus for monitoring and enhancing on-chip microprocessor reliability

• US 10,216,557 B2
• Filed: 10/23/2008
• Issued: 02/26/2019
• Est. Priority Date: 10/08/2004
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Bose, Pradip, Rivers, Jude A., Srinivasan, Jayanth
• Primary Examiner(s): Nghiem, Michael P
• Application Number:
  US12/256,907
• Publication Number:
  US 20090048808A1
• Time in Patent Office: 3,778 Days
• Field of Search: 702181, 702184, 702186
• CPC Class Codes: G06F 11/008 Reliability or availability analysis
• CPC Class Codes: G06F 11/07 Responding to the occurrence of a fault, e.g. fault tolerance

Abstract

A system and method for projecting reliability to manage system functions includes an activity module which determines activity in the system. A reliability module interacts with the activity module to determine a reliability measurement for the module in real-time based upon the activity and measured operational quantities of the system. A management module manages actions of the system based upon the reliability measurement input from the reliability module. This may be to provide corrective action, to reallocate resources, or increase reliability of the module.

Claims

1. A system for projecting reliability to manage system functions, comprising:

an activity module which determines activity in the system that occurs during operation of the system;

a reliability module interacting with the activity module to determine a reliability measurement for regions for a current period within the system in real-time based upon the activity and measured operational quantities of the system, wherein the reliability measurement characterizes one or more potential physical failure mechanisms; and

a management module comprising a processor comparing the reliability measurement within the system to a locally stored reliability target and increase activity of the system during operation of the system based on whether the reliability measurement is determined to be above or below the stored reliability target;

wherein increasing the activity of the system includes one of increasing a clock rate, reallocating resources, and increasing current or voltage.

2. The system as recited in claim 1, wherein the activity of the system is determined by an instruction per cycle analysis based on a number of instructions per cycle determined as part of the reliability measurement.

3. The system as recited in claim 1, wherein the measurement of activity is determined by counts on counters provided in the system.

4. The system as recited in claim 1, wherein the activity of the system includes at least one of temperature, power, frequency measurements and energy estimates.

5. The system as recited in claim 1, wherein the system includes a chip and the reliability module is integrated in the chip.

6. The system as recited in claim 1, wherein the system includes a chip and the management module is integrated in the chip.

7. The system as recited in claim 1, wherein the reliability module includes a software module to determine the reliability measurement.

8. The system as recited in claim 1, wherein the reliability measurement includes a mean time to failure value for the entire system.

9. The system as recited in claim 1, wherein the management module compares the reliability measurement of a current period with a reliability measurement of a previous period and takes action to increase or reduce operating conditions to adjust a reliability measurement in a subsequent period.

10. The system as recited in claim 9, wherein the action to increase or reduce operating conditions includes one of adjusting a clock rate, reallocating resources, adjusting current or voltage.

11. The system as recited in claim 1, wherein the management module employs the reliability measurement of a current period to take action to increase or reduce operating conditions to adjust a reliability measurement in a subsequent period based on a comparison of the reliability measurement of the current period to a threshold.

12. The system as recited in claim 11, wherein the action to increase or reduce operating conditions includes one of adjusting a clock rate, reallocating resources, adjusting current or voltage.

13. The system as recited in claim 1, wherein the activity module comprises a temperature sensor.

14. An integrated circuit having a reliability management system integrated therein, comprising:

an activity module which determines activity within regions of the integrated circuit that occurs during operation of the integrated circuit;

a reliability module interacting with the activity module to determine a reliability measurement for the regions for a current period within the system for the circuit in real-time based upon the activity and measured operational quantities of the integrated circuit, wherein the reliability measurement characterizes one or more potential physical failure mechanisms; and

a management module comprising a processor comparing the reliability measurement within the system to a locally stored reliability target and increase activity of the integrated system during operation of the integrated circuit based on whether the reliability measurement is determined to be above or below the stored reliability target;

wherein increasing the activity of the system includes one of increasing a clock rate, reallocating resources, and increasing current or voltage.

15. The integrated circuit as recited in claim 14, wherein the integrated circuit includes a chip and the reliability module is integrated in the chip.

16. The integrated circuit as recited in claim 14, wherein the integrated circuit includes a chip and the management module is integrated in the chip.

17. The integrated circuit as recited in claim 14, wherein the reliability measurement includes a mean time to failure value.