Service grid can provide us an environment to distribute tasks to executors around the world. Since there exist time zone difference and different working calendar systems for independent organizations, there will be some organizations are on-duty while others are off work at a specified time point. Therefore, the time factors can play an important role for job optimization scheduling in service grid. Aiming at utilizing the time difference, this paper proposed and constructed a mathematical model to solve the problem of completion time optimization of task scheduling in these environments. A case study is analyzed and it is shown that the whole execution time of the tasks can be shortened a lot while compared with the local execution scheduling model.

Aimed at the present maintenance training in the army, the virtual maintenance training system frame of a certain type of power station was designed by using the object model and virtual reality technology. The key technologies to build simulation models based on NGrain were studied, and the virtual maintenance training system was developed finally. Results show that the system has lots of advantages such as multiple training modes, friendly operation and low requests to hardware.