This study aimed to investigate performance of ARM-based computer clusters using two-phase immersion cooling approach, and demonstrate its potential benefits over the air-based natural and forced convection approaches. Clusters were created using Raspberry Pi, a commodity-level, single-board computer.
Experiments involved running benchmarking tests Sysbench, High Performance Linpack, and the combination of both, to quantify the key parameters of device junction temperature, frequency, execution time, computing performance, and energy consumption. Immersion cooling mode utilized two types of dielectric liquids, HFE-7000 and HFE-7100.
Results indicated that immersion cooling achieves up to 35% enhancement in computing performance, 40% reduction in execution time, and 150% saving in energy consumption compared to baseline natural convection. This approach is particularly beneficial for high performance computing applications where conventional air-based cooling methods would fail.
For further information contact Dr. Huseyin Bostanci.