Current students

Modern applications require the elaboration of massive amounts of data. Often, due to the computational power required, such applications may execute in data centers that consume immense amounts of energy. In 2020, data centers contributed to 2% of the world's CO2 emissions with an increasing trend. In such cases, designers must guarantee not only a high quality of the result but also efficiently manage the energy required by the computation to reduce costs and CO2 production. To achieve high performance and power savings, many data centers are moving towards heterogeneous architectures equipped with specialized hardware. Thanks to customization, these architectures can significantly minimize energy consumption, while hardware parallelism can optimize the execution time. However, such components have limited flexibility. Once they are designed, they cannot execute the functionality differently. Also, the energy consumption is fixed and depends on the implementation of the architecture. This research proposes to implement an adaptability system on the emergent FPGA to guarantee flexibility, develop different versions of a computation component, and select which version is better used at run time. In this way, based on the stimuli coming from the environment, such as the intensity of the incoming traffic or the data formats, it will be possible to use logic with different energy profiles. This approach not only allows to limit energy consumption within data centers but also to always perform the best solution according to external stimuli.