Speeding up Smartphone-based Dew Computing In-vivo Experiments Setup Via an Evolutionary Algorithm [JCR]

VIRGINIA YANNIBELLI; MATÍAS HIRSCH; JUAN TOLOZA; MAJCHRZAK, TIM A.; ALEJANDRO ZUNINO; CRISTIAN MATEOS

SENSORS

MOLECULAR DIVERSITY PRESERVATION INTERNATIONAL-MDPI

Lugar: Basel; Año: 2023

1424-8220

Dew computing aims at minimizing the dependency on remote clouds by exploiting nearby nodes for solving non-trivial computational tasks, e.g. AI inferences. Nowadays, smartphones are good candidates for rendering service as computing nodes; hence, smartphone clusters have been proposed to accomplish this task and load balancing is frequently subject of in-lab research. Using real testbeds to evaluate different load balancing strategies based on energy utilization is challenging and time consuming. In principle, test repetition requires a platform to control the charging periods between repetitions. Our Motrol hard-soft device has such capability; however, it lacks a mechanism to assure and reduce the time in which all smartphone batteries reach the level required by the next test. We propose an evolutionary algorithm to execute smartphone battery (dis)charging plans to minimize test preparation time. Charging plans proposed by the algorithm might include charging at different speeds, which is achieved by charging at maximum speed while exercising energy hungry components (CPU, screen). To evaluate the algorithm, we use charging/discharging battery traces of real smartphones, and compare the time a plan takes to synchronously prepare a set of smartphones versus that of charging all smartphones at maximum speed with no resource usage.