Thermal-aware Load-distribution for Real-time Server Farms

Today’s CPUs consume a significant amount of power and generate a high level of heat, requiring an active cooling system to support reliable operations. These CPUs avoid thermal breakdowns–in case of cooling system failures–through frequency-throttling mechanisms at the expense of a large performance penalty. In this paper, we first develop and validate a systemwide thermal model that can account for various thermal fault types such as failure of a CPU fan, faults in the case fan, and air-conditioning malfunctions. This thermal model has been validated through experimentation and measurements in AMD Linux boxes and it is incorporated into a web-server farm simulator. We use the thermal model as a temperature prediction mechanism to influence the load-distribution algorithm to minimize the impact of the thermal faults to the data center QoS. Our results show that our temperature-aware load distribution algorithm successfully keeps CPU temperatures within the desired thermal envelope, even in the presence of faults.