Summary form only given: Networked sensors and actuators are increasingly pervasive in our daily lives, from tracking inventory in warehouses to monitoring movement of environmental pollutants and helping elderly living a more independent life. The data collected by the sensors needs to be distilled into higher level information and knowledge that must be promptly acted upon by autonomous controllers or humans. The scale of these systems-sometimes called Internet of Things (IOT), Cyber-Physical Systems (CPS), or Sensor Networks (Sensornet)-dwarfs the Internet as we know today. In addition to the challenges of making the sensors small, affordable, and energy efficient, and networking them into a reliable ensemble, the more arduous task is the timely processing of voluminous and streaming sensor data subject to energy and network constraints and uncertainties. To address these challenges, we must leverage the advances in disciplines such as devices, wireless communication, control and decision theory, machine learning, and database, to name a few. Cloud computing provides new opportunities in aggregating sensor data and exploiting the aggregates for greater coverage and relevancy, and yet at the same time exacerbates the issue of privacy and security. A multi-disciplinary, holistic approach is needed.