Machine learning guided thermal management of Open Computing Language applications on CPU‐GPU based embedded platforms

As embedded devices start supporting heterogeneous processing cores (Central Processing Unit [CPU]–Graphical [GPU] based cores), performance aware task allocation becomes a major issue. Use of Open Computing Language (OpenCL) applications on both CPU and GPU improves resolves the problem. However, it has an adverse effect overall power consumption operating temperature system. Operating kind within small form factor at high frequency causes rise in which turn leads to increase processor temperature. The elevated brings about thermal issues. In this paper, we present our investigation role during execution specific application argue against running frequency. addition, machine learning guided mechanism predict optimal OpenCL kernels is presented study. Our experiments with state art ODROID XU4 platform show that experimental board if operated proposed by Machine Learning-based predictive method 12.5°C reduction 1.06% degradation compared baseline (default governor platform).