Abstract: A machine learning system includes a host device and several client devices. The client devices receive a host model from the host device, respectively, and include a first and a second client devices. The first and the second client devices store a first and a second parameter sets, respectively, and perform training on the received host models according to the first and the second parameter sets, respectively, to respectively generate a first and a second training results. If the host device has received the first training result corresponding to an m-th training round but has not received the second training result corresponding to a n-th round training, when a difference between m and n is not higher than a threshold value, the host device updates the host model according to the first training result without using the second training result.

Abstract: A server and a cloud computing resource optimization method thereof for big data cloud computing architecture are provided. The server runs a dynamic scaling system to perform the following operations: receiving a task message; executing a profiling procedure to generate a profile based on an to-be-executed task recorded in the task message; executing a classifying procedure to determine a task classification of the to-be-executed task; executing a prediction procedure to obtain a plurality of predicted execution times corresponding to a plurality of computing node numbers, a computing node type and a system parameter of the to-be-executed task; executing an optimization procedure to determine a practical computing node number of the to-be-executed task; and transmitting an optimization output message to a management server to make the management server allocate at least one data computing system to execute a program file of the to-be-executed task.

Abstract: A server and a cloud computing resource optimization method thereof for cloud big data computing architecture are provided. The server runs a dynamic scaling system to perform the following operations: receiving a task message; executing a profiling procedure to generate a profile based on an to-be-executed task recorded in the task message; executing a classifying procedure to determine a task classification of the to-be-executed task; executing a prediction procedure to obtain a plurality of predicted execution times corresponding to a plurality of computing node numbers, a computing node type and a system parameter of the to-be-executed task; executing an optimization procedure to determine a practical computing node number of the to-be-executed task; and transmitting an optimization output message to a management server to make the management server allocate at least one data computing system to execute a program file of the to-be-executed task.