Providing a computing system with real-time capabilities

• US 8,230,446 B2
• Filed: 11/28/2007
• Issued: 07/24/2012
• Est. Priority Date: 11/28/2007
• Status: Expired due to Fees

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: McKenney, Paul E., Salzberg, Claudia, Vandenbroeck, Rene, Whetzel, John K., Reynolds, Peter H., Asselin, Albert A., Mannthey, Keith, Smith, Torez, Franke, Jeffrey, Tso, Theodore Y.
• Primary Examiner(s): Sough, S.
• Assistant Examiner(s): DORAIS, CRAIG C
• Application Number:
  US11/946,511
• Publication Number:
  US 20090138896A1
• Time in Patent Office: 1,700 Days
• Field of Search: None
• US Class Current: 719/318
• CPC Class Codes: G06F 9/4812 by interrupt, e.g. masked
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

A computing system is provided with real-time capabilities so that the system is capable of running applications such that one or more real-time criteria are satisfied. An interrupt architecture of the computing system is disabled. The interrupt architecture generates interrupts sent to a firmware of the computing system in response to events. A different architecture is substituted within the computing system for the interrupt architecture. The different architecture is responsive to the events without violating the real-time criteria. In response to the events occurring, the different architecture causes one or more corrective actions to be performed.

Claims

1. A method for providing a computing system with real-time capabilities so that the computing system is capable of running applications such that one or more real-time criteria are satisfied, comprising:

disabling an interrupt architecture of the computing system while still permitting other code of the computing system to run, the interrupt architecture generating interrupts sent to a firmware of the computing system in response to events;

substituting a different architecture within the computing system for the interrupt architecture of the computing system, the different architecture responsive to the events, without violating the real-time criteria that govern the real-time capabilities, the real-time criteria comprising that the other code has to be executed such that the other code responds to other events within a time threshold; and

,

in response to the events occurring, the different architecture of the computing system causing one or more corrective actions to be performed without generating any interrupts,

wherein the other code is code other than the interrupt architecture and the different architecture,

wherein the other events are events other than the events in response to which the interrupt architecture generates the interrupts,

wherein the other code does not process interrupts,

and wherein the method provides for overcoming any potential for interrupts to affect an ability of the other code from satisfying the real-time criteria that otherwise exists.

2. The method of claim 1, wherein the events are selected as one or more of thermal events of the computing system, power consumption events of the computing system, and error events of the computing system.

3. The method of claim 2, wherein one or more of:

the thermal events comprise an operating temperature of the computing system being greater than a threshold temperature;

the power consumption events comprise a power consumption of the computing system being greater than a threshold power consumption; and

,

the error events comprise single-bit memory errors having been detected.

4. The method of claim 1, wherein the real-time criteria specify that the applications executed on the computing system are able to respond to requests in no more than a predetermined amount of time.

5. The method of claim 1, wherein disabling the interrupt architecture of the computing system comprises:

disabling non-maskable interrupts that are sent to the firmware in response to the events occurring.

6. The method of claim 5, wherein disabling the non-maskable interrupts comprises:

disabling system management interrupts (SMI'"'"'s) that are sent to the firmware in response to the events occurring.

7. The method of claim 1, wherein substituting the different architecture for the interrupt architecture comprises modifying an operating system of the computing system so that the events are monitored in a preemptible manner such that the real-time criteria are not violated and such that interrupts are not employed.

8. The method of claim 1, wherein substituting the different architecture for the interrupt architecture comprises providing a diagnostic utility running on the computing system so that the events are monitored in a preemptible manner such that the real-time criteria are not violated and such that interrupts are not employed.

9. The method of claim 1, wherein substituting the different architecture for the interrupt architecture comprises:

providing an interface between an operating system of the computing system and a diagnostic utility running on the computing system so that a driver of the operating system collects and translates the events to a format understandable by the firmware of the computing system.

10. The method of claim 1, wherein the different architecture of the computing system causing the corrective actions to be performed in response to the events occurring comprises one or more of:

causing one or more processors of the computing system to be run at a less-than-maximum clock frequency that reduces power consumption of the computing system and that reduces operating temperature of the computing system without violating the real-time criteria; and

,

where the computing system comprises a plurality of computing devices installed in a shared chassis, shutting down one or more of the computing devices to reduce power consumption of the computing system and to reduce operating temperature of the computing system without violating the real-time criteria.

11. The method of claim 1, wherein the different architecture of the computing system causing the corrective actions to be performed in response to the events occurring comprises, where the computing system comprises a plurality of computing devices, one or more of:

migrating one or more of the applications running on the computing system from one or more of the computing devices that no longer have at least one of redundant power supply capabilities and redundant cooling capabilities so that the real-time criteria will not be violated in relation to the applications; and

,

migrating one or more of the applications running on the computing system from one or more of the computing devices that are under at least one of thermal stress and power consumption stress so that the real-time criteria are not violated in relation to the applications.

12. The method of claim 1, further comprising:

modifying the firmware so that the firmware is able to detect when the operating system is no longer operating in a real-time mode, to permit the firmware to conclude that the real-time capabilities of the computing system are no longer required.

13. The method of claim 1, further comprising:

providing an interface by which an operating system of the computing system is able to notify the firmware that the computing system is to be operated in a real-time mode such that that the real-time criteria are to be satisfied; and

,

providing an interface by which an operating system of the computing system is able to notify the firmware that the computing system is no longer to be operated in the real-time mode such that the real-time criteria no longer have to be satisfied,

wherein the interrupt architecture of the computing system is disabled by the firmware in response to the operating system notifying the firmware that the computing system is to be operated in the real-time mode, and

wherein the interrupt architecture of the computing system is enabled by the firmware in response to the operating system notifying the firmware that the computing system is no longer to be operated in the real-time mode.

14. A computing system comprising:

one or more applications to be run on the computing system in real-time such that one or more real-time criteria are satisfied, the real-time criteria comprising that the one or more applications have to be executed such that the one or more applications respond to other events within a time threshold;

firmware;

an interrupt architecture by which generated interrupts are sent to the firmware in response to events; and

,

an additional architecture responsive to the events without violating the real-time criteria and without generating any interrupts,

wherein the interrupt architecture is disabled without disabling the one or more applications of the computing system to run and the additional architecture is enabled to run the applications on the computing system in real-time such that the real-time criteria are satisfied,

wherein the other code is code other than the interrupt architecture and the different additional architecture,

wherein the other events are events other than the events in response to which the interrupt architecture generates the interrupts,

wherein the other code does not process interrupts,

and wherein the computing system provides for overcoming any potential for interrupts to affect an ability of the other code from satisfying the real-time criteria that otherwise exists.

15. The computing system of claim 14, wherein the additional architecture comprises one or more of:

a diagnostic utility; and

,

modifications to an operating system of the computing system,

such that one or more of the diagnostic utility and the operating system monitor the events in a preemptible manner such that the real-time criteria are not violated and such that interrupts are not employed.

16. The computing system of claim 14, wherein the additional architecture comprises:

a first interface by which an operating system of the computing system is able to notify the firmware that the computing system is to be operated in a real-time mode such that the real-time criteria are to be satisfied; and

,

a second interface by which an operating system of the computing system is able to notify the firmware that the computing system is no longer to be operated in the real-time mode such that the real-time criteria no longer have to be satisfied.

17. The computing system of claim 14, wherein the additional architecture interacts with a management module to cause one or more corrective actions to be performed in response to the events occurring.

18. The computing system of claim 17, wherein the corrective actions comprise one or more of:

causing one or more processors of the computing system to be run at a less-than-maximum clock frequency that reduces power consumption of the computing system and that reduces operating temperature of the computing system without violating the real-time criteria;

where the computing system comprises a plurality of computing devices installed in a shared chassis, shutting down one or more of the computing devices to reduce power consumption of the computing system and to reduce operating temperature of the computing system without violating the real-time criteria;

where the computing system comprises the plurality of computing devices, migrating one or more of the applications from one or more of the computing devices that no longer have at least one of redundant power supply capabilities and redundant cooling capabilities so that the real-time criteria will not be violated in relation to the applications; and

,

where the computing system comprises the plurality of computing devices, migrating one or more of the applications from one or more of the computing devices that are under at least one of thermal stress and power consumption stress so that the real-time criteria are not violated in relation to the applications.

19. An article of manufacture comprising:

a recordable data storage medium; and

,

means in the medium for providing a computing system with real-time capabilities so that the computing system is capable of running applications such that one or more real-time criteria are satisfied, the real-time criteria comprising that other code has to be executed such that the other code responds to other events within a time threshold, by;

disabling an interrupt architecture of the computing system while still permitting the other code to run, the interrupt architecture generating interrupts sent to a firmware of the computing system in response to events; and

,

substituting a different architecture within the computing system for the interrupt architecture of the computing system, the different architecture responsive to the events without generating any interrupts,

wherein the other code is code other than the interrupt architecture and the different architecture,

wherein the other events are events other than the events in response to which the interrupt architecture generates the interrupts,

wherein the other code does not process interrupts,

and wherein the article provides for overcoming any potential for interrupts to affect an ability of the other code from satisfying the real-time criteria that otherwise exists.

20. The article of manufacture of claim 19, wherein substituting the different architecture for the interrupt architecture comprises one or more of:

performing at least one of modifying an operating system of the computing system and providing a diagnostic utility running on the computing system so that the events are monitored in a preemptible manner such that the real-time criteria are not violated and such that interrupts are not employed; and

,

providing an interface between the operating system and the diagnostic utility running on the computing system so that a driver of the operating system collects and translates the events to a format understandable by the firmware of the computing system.