GPU: Power vs Performance

GPUs are widely being used to meet the ever increasing demands of High performance computing. High-end GPUs are one of the highest consumers of power in a computer. Power dissipation has always been a major concern area for computer architects. Due to power efficiency demands modern CPUs have moved towards multicore architectures. GPUs are already massively parallel architectures. There has been some encouraging results for power efficiency in CPUs by applying DVFS . The vision is that a similar approach would also provide encouraging results for power efficiency in GPUs. In this thesis we have analyzed the power and performance characteristics of GPU at different frequencies. To help us achieve that, we have made a set of microbenchmarks with different levels of memory boundedness and threads. We have also used benchmarks from CUDA SDK and Parboil. We have used a GTX580 Fermi based GPU. We have also made a hardware infrastructure that accurately measures the power being consumed by the GPU