Parallel distributed processing system and method

PURPOSE: A distributed parallel processing system of massive data and a method thereof are provided to process large-scale and massive data, which is collected and stored, in parallel at high speed, thereby processing the large-scale data at high speed. CONSTITUTION: An expansion job tracker receives a task request from a client. The expansion job tracker inquires whether or not a task is defined in a GPU table. If the task is not defined in the GPU table, a map-reduce task is allocated in a task tracker. If the task is defined in the GPU table, the map-reduce task is allocated in a GPU task tracker. [Reference numerals] (AA) Hadoop; (BB) Extended job tracker; (CC) Task tracker; (DD) Map/reduce; (EE) GPU task tracker; (FF) Data handler; (GG) GPS map/reduce; (HH) Data node