Predict computing platform memory power utilization

• US 7,752,468 B2
• Filed: 06/06/2006
• Issued: 07/06/2010
• Est. Priority Date: 06/06/2006
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Ambrose, Anthony, Mukherjee, Udayan, Hiremath, Chetan, Chandwani, Neelam, Dodeja, Rakesh
• Primary Examiner(s): CHEN, TSE W
• Application Number:
  US11/447,701
• Publication Number:
  US 20070283178A1
• Time in Patent Office: 1,491 Days
• Field of Search: 713/1, 713/300
• US Class Current: 713/300
• CPC Class Codes: G06F 1/3203 Power management, i.e. event-based initiation of a power-saving mode
• CPC Class Codes: G06F 1/3225 of memory devices

Abstract

A method is to include implementing at least one statistical prediction model to predict memory power utilization and reduce power consumption for a computing platform. The implementation includes determining a configuration parameter for the computing platform, monitoring an operating parameter for the computing platform and predicting memory power utilization for the computing platform based on the determined configuration parameter and the monitored operating parameter. The method is to also include transitioning at least one memory module resident on the computing platform to one of a plurality of power states based at least in part on memory power utilization predicted via the implementation of the at least one statistical prediction model.

Claims

1. A method comprising:

implementing at least one statistical prediction model to predict memory power utilization and reduce power consumption for a computing platform, implementation to include;

determining a configuration parameter for the computing platform by at least identifying resources present on the computing platform;

identifying an operating parameter associated with at least one of the resources present on the computing platform;

monitoring the operating parameter of the computing platform during runtime of the computing platform by inspecting at least one of a memory table or a memory register that is updated by the resource of the computing platform during runtime of the computing platform; and

predicting memory power utilization for the computing platform during the runtime of the computing platform based on the determined configuration parameter and the monitored operating parameter; and

transitioning a current power state of at least one memory module resident on the computing platform to one of a plurality of power states, wherein transitioning the current power state is based at least in part on a statistical prediction of memory power utilization determined from the configuration parameter and the monitored operating parameter.

2. A method according to claim 1, wherein the at least one statistical prediction model includes using a statistical parameter in a prediction algorithm that is determined during a given period of time that begins as the computing platform is initially powered-up.

3. A method according to claim 1, wherein the at least one statistical prediction model includes using a statistical parameter in a prediction algorithm that is determined during a given period of time and tuned periodically during the runtime of the computing platform.

4. A method according to claim 3, wherein the statistical parameter is to include at least one statistical parameter selected from the following group of:

memory requests made to a memory controller for the at least one memory module, processing element utilization for a processing element on the computing platform, power state of a processing element on the computing platform, memory capacity utilization for a processing element on the computing platform and network bandwidth on at least one communication link between the computing platform and a network.

5. A method according to claim 4, wherein the power state of the processing element includes one of a suspend power state, a standby power state and a deep sleep power state.

6. A method according to claim 1, wherein determining the configuration parameter comprises obtaining the configuration parameter from a memory controller for the at least one memory module, the configuration parameter to include at least one usage configuration for the at least one memory module selected from the following group of:

memory interleaving, memory mirroring, memory sparing and rank order allocation.

7. A method according to claim 6, wherein monitoring the operating parameter comprises monitoring the operating parameter by obtaining information from at least one memory register at one of a processing element, the memory controller, a network interface and the at least one memory module.

8. A method according to claim 7, wherein obtaining information from the at least one memory register at the processing element comprises the information to include at least one selected from the following group of:

processing element utilization, processing element performance and a processing element power state.

9. A method according to claim 1, wherein transitioning the at least one memory module to one of the plurality of power states comprises the power states to include an offline state, an online state, a standby state and a suspend state.

10. A method according to claim 9, wherein the at least one memory module is a fully buffered dual inline memory module (FB-DIMM).

11. A method according to claim 10, wherein transitioning the FB-DIMM to one of the plurality of power states results in another FB-DIMM resident on the computing platform also transitioning to the same power state.

12. A machine-accessible storage medium comprising content, which, when executed by a machine resident on a computing platform causes the machine to:

implement at least one statistical prediction model to predict memory power utilization and reduce power consumption for a computing platform, implementation to include;

determining a configuration parameter for the computing platform by at least identifying resources present on the computing platform;

identifying an operating parameter associated with at least one of the resources present on the computing platform;

monitoring the operating parameter of the computing platform during runtime of the computing platform by inspecting at least one of a memory table or a memory register that is updated by the resource of the computing platform during runtime of the computing platform; and

predicting memory power utilization for the computing platform based on the determined configuration parameter and the monitored operating parameter during the runtime of the computing platform; and

transition a current power state of at least one memory module resident on the computing platform to one of a plurality of power states, wherein transitioning the current power state is based at least in part on a statistical prediction of memory power utilization determined from the configuration parameter and the monitored operating parameter.

13. A machine-accessible storage medium according to claim 12, wherein determining the configuration parameter comprises obtaining the configuration parameter from a memory controller for the at least one memory module, the configuration parameter to include at least one usage configuration for the at least one memory module selected from the following group of:

memory interleaving, memory mirroring, memory sparing and rank order allocation.

14. A machine-accessible storage medium according to claim 13, wherein monitoring the operating parameter comprises monitoring the operating parameter by obtaining information from at least one memory register at one of a processing element, the memory controller, a network interface and the at least one memory module.

15. A machine-accessible storage medium according to claim 14, wherein obtaining information from the at least one memory register at the processing element comprises the information to include at least one selected from the following group of:

processing element utilization, processing element performance and a processing element power state.