Automated firmware settings verification

• US 9,471,784 B1
• Filed: 12/06/2012
• Issued: 10/18/2016
• Est. Priority Date: 12/06/2012
• Status: Active Grant

• Current Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Original Assignee: Amazon Technologies, Inc. (Amazon.com, Inc.)
• Inventors: Marr, Michael David, Aithal, Anirudh Balachandra, Klein, Matthew David
• Primary Examiner(s): Huynh, Kim
• Assistant Examiner(s): Kinsey, Brandon
• Application Number:
  US13/706,839
• Time in Patent Office: 1,412 Days
• US Class Current: 1/1
• CPC Class Codes: G06F 21/572 Secure firmware programming, e.g. of basic input output system [BIOS]

Abstract

Systems and methods are described for managing computing resources. In one embodiment, data representative of an abstracted firmware framework is maintained. The data may comprise computing firmware settings and determined based on standardized associations between vendor-specific firmware settings and abstracted firmware settings that are independent of the vendor-specific firmware settings. In response to receiving a request for a computing firmware setting, the requested computing firmware setting is translated to one or more vendor-specific firmware settings based on the data. A computing resource capable of implementing the one or more vendor-specific firmware settings is identified.

Claims

1. A computing system comprising at least one computing node and at least one data store in communication with the at least one computing node, the at least one data store having stored therein computer instructions, that upon execution by the at least one computing node, cause at least:

forming groupings of computing devices in a computing environment providing computing resources, the groupings having common computing attributes corresponding to one or more firmware settings of an abstraction firmware framework, the abstraction firmware framework having an associated interface for mapping relationships between hardware-specific firmware settings for a plurality of the computing devices and a plurality of selectable abstracted firmware settings that are translatable to the hardware-specific firmware settings, wherein the abstracted firmware settings are different from the hardware-specific firmware settings, wherein the abstracted firmware settings are not vendor-specific, and the abstracted firmware settings correspond to one or more desired computing attributes of the plurality of computing devices;

receiving a request for a selected abstracted firmware setting;

determining which of the groupings can support the selected abstracted firmware setting; and

creating a new grouping when it is determined that none of the groupings can support the selected abstracted firmware setting, wherein the new grouping is based in part on verification that a computing device in the new grouping that has implemented the selected abstracted firmware setting meets or exceeds at least one performance criterion.

2. The computing system according to claim 1, wherein the groupings are based on at least one computing environment management criterion.

3. The computing system according to claim 2, wherein the at least one computing environment management criterion is a firmware grouping policy.

4. The computing system according to claim 1, further comprising implementing an expert system configured to provide a decision-making capability for forming the groupings.

5. A computer-implemented method for managing computing resources, comprising:

maintaining data representative of an abstracted firmware framework comprising computing firmware settings, the data determined based on standardized associations performed using an interface associated with the abstracted firmware framework, the associations between hardware-specific firmware settings of a plurality of the computing resources and a plurality of abstracted firmware settings that are independent of the hardware-specific firmware settings, wherein the abstracted firmware settings are different from the hardware-specific firmware settings, wherein the abstracted firmware settings are independent of the hardware-specific settings, and wherein the abstracted firmware settings correspond to one or more desired computing attributes;

verifying at least one of the hardware-specific firmware settings to confirm that it meets predetermined criteria;

receiving a request for a selected abstract firmware setting;

translating the requested selected abstract firmware setting to one or more vendor-specific firmware settings based on the data; and

identifying a computing resource capable of implementing the one or more vendor-specific firmware settings.

6. The method of claim 5 wherein the hardware specific firmware settings include at least one of a basic input/output system (BIOS), a non-uniform memory access (NUMA), processor clock rate, processor clock frequency scaling, performance state or power state setting.

7. The method of claim 5 further comprising maintaining groupings of computing resources, each of the groupings comprising one or more computing resources that have a common set of computing firmware settings that have been incorporated and verified.

8. The method of claim 5 wherein the predetermined criteria includes a data center capacity management policy.

9. The method of claim 7, further comprising sending an indication to a source of the request as to which, if any, of the groupings have incorporated and verified the computing firmware settings.

10. The method of claim 8 wherein the data center capacity management policy is a target availability goal.

11. The method of claim 5 further comprising causing identification of at least one computing resource to incorporate and verify the selected abstract firmware setting and initiating at least one verification task to verify that the computing resource satisfies the predetermined criteria after incorporating the selected abstract firmware setting.

12. The method of claim 11 further comprising evaluating data received from the at least one computing resource and verifying that the computing resource satisfies the predetermined criteria after incorporating the selected abstract firmware setting.

13. The method of claim 5, further comprising using a fitness function to determine whether the at least one of the hardware-specific firmware settings meets the predetermined criteria.

14. One or more non-transitory computer-readable storage media having collectively stored thereon executable instructions that, when executed by one or more processors of a computer system, cause the computer system to:

receive a selected abstract firmware setting; and

identify at least one computing resource capable of implementing firmware settings corresponding to the selected abstract firmware setting, the at least one computing resource identified based on a determination that one or more hardware-specific firmware settings incorporated on the at least one computing resource is compatible with the selected abstract firmware setting, wherein it is verified that the at least one computing resource satisfies predetermined criteria after implementation of the firmware settings, and wherein;

a plurality of abstract firmware settings are mapped using an interface of an abstraction firmware framework to a plurality of hardware-specific firmware settings for a plurality of computing resources;

the abstract firmware settings are different from the hardware-specific firmware settings, and

the abstract firmware settings are independent of vendor-specific settings.

15. The computer-readable storage media of claim 14 further storing thereon executable instructions that, when executed by the one or more processors of a computer system, cause the computer system to adjust the firmware settings based on a fitness function.

16. The computer-readable storage media of claim 15 further storing thereon executable instructions that, when executed by the one or more processors of a computer system, cause the computer system to use at least one criterion as a performance goal.

17. The computer-readable storage media of claim 14 wherein performance testing is performed iteratively to identify a final set of firmware settings based on a fitness function.