Effect of voltage scaling on symmetric multicore's speed-up

As the technology scales further, multicore and many-core processors emerged as an alternative to keep up with performance demands. But because of power and thermal constraints we are obliged to power off remarkable area of chip. Lots of ideas have been presented to increase energy efficiency and make utilization as high as possible. In this paper, we discuss different models and methods of exploiting dark silicon, and by using dynamic voltage and frequency scaling in Amdahl's law, we attempt to decrease the amount of dark silicon and improve total performance. We propose the best symmetric multicore model for variety of parallelism in workloads at different power budget constraints. According to the results, by voltage scaling, we reach approximately 6.43× improvement in speedup for a high parallel workload.