Application of an Improved Immune Evolutionary Algorithm to Grid Task Scheduling
ShengJun Xue,Fangfang Liu
DOI: https://doi.org/10.4028/www.scientific.net/amr.219-220.1383
2011-01-01
Advanced Materials Research
Abstract:Grid task scheduling is an NP problem, performance of scheduling algorithms greatly influences scheduling results. Aiming at the shortages of the existing Evolutionary Algorithm, such as premature convergence, search process easily falling into local optimum, poor scheduling results and so on, this paper proposed an improved immune Evolutionary Algorithm which introduced concentration mechanism in the immune system into Immune Evolutionary Algorithm and adjusted regulator to adaptive function. Simulation experiment shows that, convergence speed and performance of the improved algorithm are significantly improved and it can better converge to global optimal solution, applying the algorithm to grid task scheduling can obtain better scheduling results.