| 摘要 |
The present invention provides a method and system for grid computing. In an embodiment, a plurality of client machines are interconnected to at least one master machine. The master machine assigns a portion of a computing task to each one of the client machines. If any given client machine fails, or is delayed, in the performance its portion of the task, the master machine uses an estimate of that particular portion when presenting output f or the task.
|
