In order to eliminate the current negative condition of Automatic Computer Room Air-Conditioning (CRAC) system, self-tuning Fuzzy Logic Control (FLC) was designed and applied to fan speed in CRAC system. In this paper, we derive a thermodynamic model of a datacenter suitable for applying adaptive self-tuning PID-type fuzzy adaptive control theory. It combines the classic PID control strategy and fuzzy adaptive control theory. The classic PID control uses the error and rate of change of error as its inputs to control the temperature automatically, and the fuzzy logic controller is used in the self-tuning PID-type fuzzy control to tune the parameters of PID controller on-line by fuzzy control rules. Simulation and testing results show that the proposed self-tuning FLC method can achieve less steady-state error and short settling time in temperature control of datacenter.