Beijing Climate Centre (BCC) of China Meteorological Administration (CMA) has recently developed and completed a new-generation Climate Intelligent Prediction and Analysis System (CIPAS2.3), which is a crowdsourcing innovative platform based on the big data technology. Based on China Integrated Meteorological Information Service System (CIMISS), the data application environment of CIPAS2.3 containing observed, climate index, reanalysis, satellite and numerical forecast data and products, as well as data collected uniformly with the Network Common Data Form (NetCDF), is deployed on the cloud sharing platform. With the new technology architecture and development process to using, CIPAS2.3 separates "operation" and "technology" while integrating the data file library, running platform and algorithm library functions, to realize the unification of data, climate algorithm and climate prediction product. In CIPAS2.3, all operation processes have been customized by using process scheduling techniques, and with the “custom configuration” function, which support the new business process added, to realize the crowdsourcing innovation of the operation process. The algorithm library has strict access specification for all kinds of climate algorithm, as long as the algorithm comply with the Standard Specification for development, registration and invoking, it can monitor the whole process of the algorithm library. CIPAS2.3 realizes the unified deployment based on big data, meanwhile, through a unified algorithm integration platform, it provides a "climatic crowdsourcing innovation" platform for scientific research and operational personnel.