Tracking Thermal Mini-Cycle Stress

• US 20100049466A1
• Filed: 08/20/2008
• Published: 02/25/2010
• Est. Priority Date: 08/20/2008
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Floyd, Michael S., Casey, Jon A., Ware, Malcom S., Muncy, Jennifer V., Weekly, Roger D., Ghiasi, Soraya, Marston, Kenneth C.
• Granted Patent: US 7,917,328 B2
• Time in Patent Office: Days
• US Class Current: 702/130
• CPC Class Codes: G06F 11/3058 Monitoring arrangements for monitoring environmental properties or parameters of the computing system or of the computing system component, e.g. monitoring of power, currents, temperature, humidity, position, vibrations thermal management in cooling arrangements of a computing system G06F1/206

Abstract

Monitoring temperature excursions an assembly experiences over a life of the assembly is provided. A determination is made as to whether the assembly has been in service beyond a predetermined end of life objective. Responsive to the assembly failing to be in service beyond the predetermined end of life objective, a new temperature value associated with the assembly is read. A modifier value for a figure of merit (FOM) value is computed and added to a cumulative figure of merit value. The cumulative figure of merit value is compared to a cumulative stress figure of merit budget. Responsive to the cumulative figure of merit value exceeding the cumulative stress figure of merit budget, an identified stress management solution is implemented.

Claims

1. A method, in a data processing system, for monitoring temperature excursions an assembly experiences over a life of the assembly, the method comprising:

determining if the assembly has been in service beyond a predetermined end of life objective;

responsive to the assembly failing to be in service beyond the predetermined end of life objective, reading a new temperature value associated with the assembly;

computing a modifier value for a figure of merit (FOM) value;

adding the modifier value to a cumulative figure of merit value;

comparing the cumulative figure of merit value to a cumulative stress figure of merit budget; and

responsive to the cumulative figure of merit value exceeding the cumulative stress figure of merit budget, implementing an identified stress management solution.

2. The method of claim 1, computing the modifier to the figure of merit comprises:

once the data processing system has been initialized, setting a time_then variable value to a time_now variable value;

setting a temperature then (Tthen) variable value to a temperature now (Tnow) variable value;

reading a new time from the data processing system and setting the time_now variable equal to the new time;

reading the new temperature value from the assembly and setting the Tnow variable equal to the new temperature;

calculating a time2add2start variable value, wherein the time2add2start variable value is a time to add to a duration for calculating a current thermal cycle figure of merit (FOMcycle);

determining whether a current thermal cycle is failing to contribute to a FOMcycle increase;

responsive to the current thermal cycle contributing to the FOMcycle increase, calculating the modifier value, wherein the modifier value is a FOMcycle modifier value and wherein the FOMcycle modifier value is a contribution for the current thermal cycle being evaluated;

determining if the time2add2start variable value is equal to zero; and

responsive to the time2add2start variable value equaling zero, adding the modifier value into the cumulative figure of merit value.

4. The method of claim 2, further comprising:

responsive to the time2add2start variable value failing to equal zero, adding the FOMcycle modifier value to the cumulative figure of merit value minus a virtual contribution of a portion that was added due to the time2add2start variable value.

5. The method of claim 2, wherein the virtual contribution of the portion that was added due to the time2add2start variable value is calculated using the following equation:

6. The method of claim 2, wherein the time2add2start variable value is calculated using the following equation:

7. The method of claim 2, wherein determining whether the current thermal cycle is failing to contribute to the FOMcycle increase comprises:

determining if a length of time since the new temperature last changed direction as given by the time2add2start variable value is longer than a predetermined time period or if the length of time since the new temperature last changed direction as given by the time2add2start variable value has changed direction so that the time2add2start variable value is a negative value.

8. The method of claim 2, further comprising:

responsive to the current thermal cycle failing to contribute to the FOMcycle increase, determining if the time2add2start variable value is positive or equal to zero;

responsive to the time2add2start variable value being positive or zero, setting the Tcycle_start variable value equal to the Tthen variable value.

9. The method of claim 8, further comprising:

responsive to the time2add2start variable value being negative, setting the Tcycle_start variable value equal to the Tnow variable value.

10. The method of claim 2, wherein initializing the data processing system comprises:

setting the current temperature (Tnow) variable value equal to an initial temperature read by a thermal sensor associated with the assembly, wherein the Tnow variable value is set to the initial temperature of the assembly at a time the assembly is first initialized;

setting the time_now variable value equal to a time the assembly is first initialized;

setting the current thermal cycle (Tcycle_start) temperature variable equal to Tnow;

setting the cumulative figure of merit variable value equal to zero; and

setting the FOMcycle variable value equal to zero.

11. The method of claim 1, wherein the identified stress management solution is one of a group comprising:

adding pseudo workloads during periods where there is no workload demand;

suspending entry into nap modes;

delaying resumption of process activity on a core by a small time period to effectively merge high power consuming processes without substantially impacting perceived performance;

dynamically throttling clock frequency and voltage to an average performance level that regulates mini-cycle stress at the cumulative criteria;

scheduling housekeeping tasks to processor cores in which a further temperature decline would result in a mini-cycle of significance;

or

scheduling processes over available cores and threads in order to minimize thermal excursions per a cumulative thermal mini-cycle stress budget.

12. The method of claim 1, wherein the figure of merit (FOM) value correlates to stress testing results using industry standard tests for qualifying electronic parts and wherein the industry standard tests are Accelerated Thermal Cycling (ATC) tests.

21. The method of claim 2, wherein the FOMcycle modifier value is calculated using the following equation:

3. (Canceled)

13. A computer program product comprising a computer recordable medium having a computer readable program recorded thereon, wherein the computer readable program, when executed on a computing device, causes the computing device to:

determine if the assembly has been in service beyond a predetermined end of life objective;

responsive to the assembly failing to be in service beyond the predetermined end of life objective, read a new temperature value associated with the assembly;

compute a modifier value for a figure of merit (FOM) value;

add the modifier value to a cumulative figure of merit value;

compare the cumulative figure of merit value to a cumulative stress figure of merit budget; and

responsive to the cumulative figure of merit value exceeding the cumulative stress figure of merit budget, implement an identified stress management solution.

14. The computer program product of claim 13, wherein the computer readable program to compute the modifier to the figure of merit further causes the computing device to:

once the data processing system has been initialized, set a time_then variable value to a time_now variable value;

set a temperature then (Tthen) variable value to a temperature now (Tnow) variable value;

read a new time from the data processing system and set the time_now variable equal to the new time;

read the new temperature value from the assembly and set the Tnow variable equal to the new temperature;

calculate a time2add2start variable value, wherein the time2add2start variable value is a time to add to a duration for calculating a current thermal cycle figure of merit (FOMcycle);

determine whether a current thermal cycle is failing to contribute to a FOMcycle increase;

responsive to the current thermal cycle contributing to the FOMcycle increase, calculate the modifier value, wherein the modifier value is a FOMcycle modifier value and wherein the FOMcycle modifier value is a contribution for the current thermal cycle being evaluated;

determine if the time2add2start variable value is equal to zero; and

responsive to the time2add2start variable value equaling zero, add the modifier value into the cumulative figure of merit value.

16. The computer program product of claim 14, wherein the computer readable program to determine whether the current thermal cycle is failing to contribute to the FOMcycle increase further causes the computing device to:

determine if a length of time since the new temperature last changed direction as given by the time2add2start variable value is longer than a predetermined time period or if the length of time since the new temperature last changed direction as given by the time2add2start variable value has changed direction so that the time2add2start variable value is a negative value.

22. The computer program product of claim 14, wherein the FOMcycle modifier value is calculated using the following equation:

15. (canceled)

17. An apparatus, comprising:

a processor; and

a memory coupled to the processor, wherein the memory comprises instructions which, when executed by the processor, cause the processor to;

determine if the assembly has been in service beyond a predetermined end of life objective;

responsive to the assembly failing to be in service beyond the predetermined end of life objective, read a new temperature value associated with the assembly;

compute a modifier value for a figure of merit (FOM) value;

add the modifier value to a cumulative figure of merit value;

compare the cumulative figure of merit value to a cumulative stress figure of merit budget; and

responsive to the cumulative figure of merit value exceeding the cumulative stress figure of merit budget, implement an identified stress management solution.

18. The apparatus of claim 17, wherein the instructions to compute the modifier to the figure of merit further cause the processor to:

once the data processing system has been initialized, set a time_then variable value to a time_now variable value;

set a temperature then (Tthen) variable value to a temperature now (Tnow) variable value;

read a new time from the data processing system and set the time_now variable equal to the new time;

read the new temperature value from the assembly and set the Tnow variable equal to the new temperature;

calculate a time2add2start variable value, wherein the time2add2start variable value is a time to add to a duration for calculating a current thermal cycle figure of merit (FOMcycle);

determine whether a current thermal cycle is failing to contribute to a FOMcycle increase;

responsive to the current thermal cycle contributing to the FOMcycle increase, calculate the modifier value, wherein the modifier value is a FOMcycle modifier value and wherein the FOMcycle modifier value is a contribution for the current thermal cycle being evaluated;

determine if the time2add2start variable value is equal to zero; and

responsive to the time2add2start variable value equaling zero, add the modifier value into the cumulative figure of merit value.

20. The apparatus of claim 18, wherein the instructions to determine whether the current thermal cycle is failing to contribute to the FOMcycle increase further causes the processor to:

determine if a length of time since the new temperature last changed direction as given by the time2add2start variable value is longer than a predetermined time period or if the length of time since the new temperature last changed direction as given by the time2add2start variable value has changed direction so that the time2add2start variable value is a negative value.

23. The apparatus of claim 18, wherein the FOMcycle modifier value is calculated using the following equation:

19. (canceled)