These days, many dark matter direct search experiments have been built and are accumulating data. Some of them already have reported positive excess in data which are possibly interpreted as dark matter signal. In this talk, after brief review on present status of direct search experiment, we discuss about the importance of complementarity among different direct detection experiments in discriminating WIMP models by measuring both of spin-dependent and spin-independent cross section as well as its mass. More specifically, we analyze several bolometric target stested in ROSEBUD experiment and examine their competitiveness focusing on model discrimination ability.