Power Modeling Approach for GPU Source Program

Abstract

Rapid development of information technology makes our environment become smarter and massive high performance computers are providing powerful computing for that. Graphics Processing Unit (GPU) as a typical high performance component is being widely used for both graphics and general-purpose applications. Although it can greatly improve computing power, it also delivers significant power consumption and need sufficient power supplies. To make high performance computing more sustainable, the important step is to measure it. Current power technologies for GPU have some drawbacks, such as they are not applicable for power estimation at the early stage. In this article, we present a novel power technology to correlate power consumption and the characteristics at the programmer perspective, and then to estimate power consumption of source program without prerunning. We conduct experiments on Nvidia's GT740 platform; the results show that our power model is more accurately than regression model and has an average error of 2.34% and the maximum error of 9.65%.

Fig. 1. Different Approaches of energy measurement[6]

Fig. 2. Structure of BP neural network

Fig. 3. Comparison between estimated and measured power

Fig. 4. Relationship between Rreg and the power

Fig. 5. Relationship between Os and the power

Fig. 6. Relationship between Rsmem and the power

Fig. 7. Relationship between Rgmem and the power

Fig. 8. Relationship between CTMR and the power

Fig. 9. Power comparisons under different divergence andinput size

Fig. 10. Different input affect characteristics

Fig. 11. Power error under different approaches

Table 1. Resource specified by programmer

Table 2. Hardware resource limitation