This paper develops and implements a real time demand side management system on hardware in order to control unpredictable loads in a residential house application. The control is implemented under unreliable grid conditions with a high amount of energy blackouts, and complements a higher predictive control layer targeting predictable loads. A PV-battery backup system is installed in order to replace the grid during regular and frequent energy blackout periods that occur in various developing countries. The real time controller should enable or disable the operation of unpredictable devices in a reliable and fast manner according to preset priority levels applied by the user and available energy in the system. The aim of the controller is the prevention of the occurrence of a loss of power supply while respecting the operational constraints of the installed system. The developed control is implemented on the Zebdoard, which features a ZYNQ device combining dual-core ARM Cortex-A9 processors with traditional Field Programmable Gate Array (FPGA) logic fabric. The processor is programmed using an interrupt based strategy, the inputs and outputs of the control are linked to the FPGA of the ZYNQ The simulation and implementation results show that the developed management program is highly flexible, accurate, fast, and reliable. (C) 2016 Elsevier B.V. All rights reserved.