TEES: A novel multiple criteria optimization scheme for temperature-constrained energy efficient storage

• This paper made the first attempt to address the temperature overhead in existing Energy Efficient Storage.
• Fuzzy decision making technology is introduced to achieve complexity multidimensional optimization.
• The precise online temperature predict model granted the control accuracy and performance requirements.

Existing energy saving schemes that have been developed for Energy Efficient Storage funnel I/O traffic on a few disks while allowing the rest idle. These schemes can cause long standing disks to overburden, resulting in a higher rate of disk failure and reliability degradation. In this paper, we develop a novel multiple criteria optimization scheme based on Fuzzy Decision Making theory, for the Temperature-constrained Energy Efficient Storage System called TEES. TEES aims to enforce a temperature constraint as well as performance requirements while also keeping energy consumption to a minimum. This is achieved by developing an online temperature prediction model and aggregating all the decision criteria, such as I/O performance, power consumption, estimated temperature and frequency of disk-status transition. The experimental results show that TEES is able to reduce disk temperature by 20–30% as compared with existing control methods, while obtaining comparable performance and power consumption.