The disastrous effect of flood has shown its influence ifn the past, and as a result, millions of dollars infrastructure have been shattered. Even after so much research, still there is no global ubiquitous system that can collect, store and analyze big data and generate the flood prediction results. In this paper, a social collaborative Internet of Things (IoT) based smart flood monitoring and forecasting architecture is proposed with the convergence between big data and HPC. It classifies geographical areas into a web of hexagonal for effective installation of energy efficient IoT devices. All relevant flood causing and flood preventing attributes are sensed using these IoT devices and computed by big data and HPC processing. Singular Value Decomposition (SVD) is used for attributes reduction. The K-mean clustering algorithm is used to predict the current state of flood and flood rating in any location, whereas Holt-Winter's forecasting method is used to forecast the flood. Experimental evaluation is being done on meteorological data collected by the Indian government and results indicated the effectiveness of the proposed architecture.