| 主权项 |
1. A computer system, comprising:
at least one physical machine comprising at least one virtual machine and a virtual machine manager; and a graphics processing cluster comprising a plurality of graphics processing servers, each of the graphics processing servers comprising a plurality of graphics processing units, wherein a main graphics processing server out of the graphics processing servers receives a graphics processing request provided by the virtual machine manager and the at least one virtual machine, the main graphics processing server comprising a cluster load balancer, the cluster load balancer assigning the graphics processing request to a minor graphics processing server out of the graphics processing servers according to a load scale table, and the minor graphics processing server generates a graphics processing result according to the graphics processing request and transmits the graphics processing result to the at least one virtual machine through the virtual machine manager, wherein the cluster load balancer keeps updating the load scale table and assigns the graphics processing request to the graphics processing units in the graphics processing servers according to the load scale table, and wherein the cluster load balancer comprises a class interval value and calculates the number of user connections that are performing the graphics processing request in the graphics processing cluster, and when the number of the user connections is reduced by a number greater than the class interval value, one of the graphics processing servers is turned off, and when the number of the user connections increases by a number greater than the class interval value, one of the graphics processing servers that have been turned off is turned on. |
