Recent advances in on-line social networking, sensing technologies and the Internet of Things (IoT), as well as mobile and cloud computing are: • blurring the boundaries between the physical, social and cyber worlds, • fuelling the astonishing growth of internet users in the past five years (from 1.158 billion in 2007 to 2.278 billion in 2013), and • causing an explosion of big data that is being produced at a high velocity from a myriad of cyber-physical and social sources. Cyber-Social Computing encompasses real-time extraction of high value information from social networks and millions of cyber-physical systems in the Internet of Things (IoT), as well as the development of specialized cloud-based services to support this anywhere and from any device. Despite an expanding array of business, government and scientific applications that require distilling knowledge from big cyber-social data, currently there is no easy way to manage and exploit such big data, do this in real-time, or formulate cloud services that make this possible anywhere via mobile devices. Therefore, Cyber-Social Computing requires the development of novel solutions for discovering on-line cyber-physical and social media sources, dynamically integrating such sources and their data, and analysing billions of data streams and tens of years of historical data form on-line cyber-physical and social networks anywhere and in real time. In this talk we provide an overview of joint research efforts involving prominent open source innovators towards developing IoT cloud solutions for big-data exploitation. In particular, we discuss four interrelated research projects that aim to develop an open source software platform that will help springboard IoT application development in academic research institutions and SMEs around the world. In this talk, we mainly focus on the development of IoT solutions to support dynamic and semantic-based discovery and integration of internet connected cyber-physical devices as needed by each application, techniques for stream processing and real-time aggregation/summarization of IoT data, and corresponding IoT cloud services. Next, we present on-going research in developing big data aggregations techniques, such as incremental clustering and anomaly detection for crowd sourcing. Finally, we outline a unified approach for big cyber-social data management and analysis that involves real-time aggregation/summarization as the main way of having users interacting with big data, extracting high value information, or reducing the data size to a level than more traditional data analysis solutions can be applied. We also present as case studies three of the largest cyber-social systems in the world (in terms of the number of data points and data velocity they manage) we have developed at CSIRO in the domains of digital agriculture, smart energy grids, and disaster management, and discuss show how these systems utilize real-time cyber-social data aggregation/summarization to help raise agricultural production, reduce energy consumption, and mitigate disasters by providing situation awareness.