The invention relates to a system of big-data resource scheduling and a method thereof. The system includes: a data collection module, which collects status and load information of a physical or virtual node in a resource pool, and stores the same into a database; a resource integration module, which carries out full-lifecycle management on a physical or virtual resource in the system; an API engine module, which is used for providing other components to apply for resources from the system of scheduling; a scheduling management module, which is used for managing other modules by the system ofresource scheduling; a scheduling strategy engine module, which is used for being responsible for scheduling strategy processing and execution of the system of resource scheduling; a critical-value monitoring module, which is used for monitoring on resource status and load conditions, and reporting data in real time to the scheduling strategy engine module; and a resource proxy module, which is used for encapsulating operations on physical machines and virtual machines. According to the system, resource status and utilization rate information is collected, a pre-defined scheduling strategy isutilized to complete dynamic scheduling of various resources in the entire resource pool, load distribution is optimized, and resource utilization rates are increased.