PLATFORM AS A SERVICE CLOUD SERVER AND MACHINE LEARNING DATA PROCESSING METHOD THEREOF

摘要

A PaaS cloud server receives a machine learning application resource request and determines an available logic circuit object accordingly. The PaaS cloud server creates a data compression and storing service object and pairs the data compression and storing service object and the logic circuit object. The PaaS cloud server binds the data compression and storing service object with a machine learning application, and receives a machine learning data from the machine learning application. The PaaS cloud server inputs the machine learning data into a part logic circuit of an FGPA circuit corresponding to the logic circuit object according to the pairing of the data compression and storing service object and the logic circuit object so that the part logic circuit compresses and stores the machine learning data. The PaaS cloud server returns information which relates to the compressed and stored data to the machine learning application.