To increase energy efficiency, the computer architectural approach controlling the frequency and/or the number of cores at hardware/software level, and the compression approach at algorithmic level have been proposed. In the computer architectural approach, there is a tradeoff between power consumption and execution time. That is, if we increase the frequency, the power consumption is increased while the execution time is decreased. In the image compression or information security, we can improve the performance efficiency at algorithmic level and/or derive the optimal applications parameter based on the tradeoff between the video/security quality and the energy consumption. We are researching the analysis of machine characteristics and parallel applications(multimedia/crypto/image processing algorithms) characteristics collectively, and thus improve the energy efficiency of compression using a commercial multi-core platform.