Cross-layer power optimization of program code and/or software architecture
According to a technique of code development, a plurality of subsystem power profiles for a corresponding plurality of subsystems of a host platform in a production data processing environment are received. Each of the subsystem power profiles specifies power consumption of a corresponding one of the plurality of subsystems across a range of workload traffic intensities. Based on the plurality of power profiles, source code and/or software architecture of the source code is power-optimized in a development data processing environment such that, in the production data processing environment, power consumption per unit of performance of an executable program derived from the source code is improved.
1. A method of data processing, comprising:
capturing raw subsystem power data from a production data processing environment via an application programming interface;
based on the raw subsystem power data, developing a plurality of subsystem power profiles for a corresponding plurality of subsystems of a host platform in the production data processing environment, wherein each of the subsystem power profiles specifies power consumption of a corresponding one of the plurality of subsystems across a range of workload traffic intensities; and
based on the plurality of power profiles, power-optimizing source code in a development data processing environment such that, in the production data processing environment, power consumption per unit of performance of an executable program derived from the source code is improved.
2. The method of claim 1 , further comprising generating the subsystem power profiles, wherein the generating includes:
filtering raw subsystem power data from the production data processing environment using a Kalman filter.
3. The method of claim 1 , wherein power-optimizing the source code includes power-optimizing the source code based on static code analysis in the development data processing environment.
4. The method of claim 1 , wherein power-optimizing the source code includes determining workload power consumption based on workload traffic intensity and traffic power cost.
5. The method of claim 1 , wherein power-optimizing the source code includes iteratively power-optimizing the source code based on updated subsystem power profiles.
6. The method of claim 1 , further comprising outputting a report predicting improvement in power consumption and utilization of the plurality of subsystems in the production data processing environment based on updates to the source code.
7. A computer program product, the computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions being executable by a processor, wherein, when executed, the program instructions cause the processor to perform:
capturing raw subsystem power data from a production data processing environment via an application programming interface;
based on the raw subsystem power data, developing a plurality of subsystem power profiles for a corresponding plurality of subsystems of a host platform in the production data processing environment, wherein each of the subsystem power profiles specifies power consumption of a corresponding one of the plurality of subsystems across a range of workload traffic intensities; and
based on the plurality of power profiles, power-optimizing source code in a development data processing environment such that, in the production data processing environment, power consumption per unit of performance of an executable program derived from the source code is improved.
8. The program product of claim 7 , further comprising generating the subsystem power profiles, wherein the generating includes:
filtering raw subsystem power data from the production data processing environment using a Kalman filter.
9. The program product of claim 7 , wherein power-optimizing the source code includes power-optimizing the source code based on static code analysis in the development data processing environment.
10. The program product of claim 7 , wherein power-optimizing the source code includes determining workload power consumption based on workload traffic intensity and traffic power cost.
11. The program product of claim 7 , wherein power-optimizing the source code includes iteratively power-optimizing the source code based on updated subsystem power profiles.
12. The program product of claim 7 , further comprising outputting a report predicting improvement in power consumption and utilization of the plurality of subsystems in the production data processing environment based on updates to the source code.
13. A data processing system, comprising:
a processor core; and
data storage coupled to the processor core, wherein the data storage includes program code that, when executed by the processor core, causes the processor core to perform:
capturing raw subsystem power data from a production data processing environment via an application programming interface;
based on the raw subsystem power data, developing a plurality of subsystem power profiles for a corresponding plurality of subsystems of a host platform in the production data processing environment, wherein each of the subsystem power profiles specifies power consumption of a corresponding one of the plurality of subsystems across a range of workload traffic intensities; and
based on the plurality of power profiles, power-optimizing source code in a development data processing environment such that, in the production data processing environment, power consumption per unit of performance of an executable program derived from the source code is improved.
14. The program product of claim 13 , further comprising generating the subsystem power profiles, wherein the generating includes:
filtering raw subsystem power data from the production data processing environment using a Kalman filter.
15. The program product of claim 13 , wherein power-optimizing the source code includes power-optimizing the source code based on static code analysis in the development data processing environment.
16. The program product of claim 13 , wherein power-optimizing the source code includes determining workload power consumption based on workload traffic intensity and traffic power cost.
17. The program product of claim 13 , wherein power-optimizing the source code includes iteratively power-optimizing the source code based on updated subsystem power profiles.
18. The program product of claim 13 , further comprising outputting a report predicting improvement in power consumption and utilization of the plurality of subsystems in the production data processing environment based on updates to the source code.