High-performance scheduling is critical to the achievement of application performance on the computational grid. New scheduling algorithms are in demand for addressing new concerns arising in the grid environment. One of the main phases of scheduling on a grid is related to the load balancing problem therefore having a high-performance method to deal with the load balancing problem is essential to obtain a satisfactory high-performance scheduling. This paper presents SAGE, a new high-performance method to cover the dynamic load balancing problem by means of a simulated annealing algorithm. Even though this problem has been addressed with several different approaches only one of these methods is related with simulated annealing algorithm. Preliminary results show that SAGE not only makes it possible to find a good solution to the problem (effectiveness) but also in a reasonable amount of time (efficiency).
展开▼
