Title :
A new genetic-based algorithm for reliable distributed resources
Author :
Nejadkheirallah, Mohsen ; Darvish, Kobra ; Sookhtsaraei, Reza ; Shahnazi, Mehdi
Author_Institution :
Allameh Mohaddes Noori Inst. of Higher Educ., Noor, Iran
Abstract :
In this paper, a new scheduling algorithm has been introduced based on dynamic genetic algorithm, which is more efficient in comparison with the previous similar algorithms and has more reliability. SQEFG is a new proposed algorithm, with new parameters and metrics defined, delays and response time to jobs have been decreased and the user satisfaction level has been increased. Analysis of proposed algorithm and several simulations´ results show better makeSpan in the overall processing time and more efficiency improvement in distributed systems. Hence the quality of service increases at different levels of distributed processors and the acknowledgments delay time of the processes decrease significantly. Considering the majority of parameters in the Grid environment, a new scheduling algorithm for independent jobs based on genetic algorithm has been presented which has increased the performance in the grid environments. For fault tolerance control in grid environments with high failure rates, inspection point method that has an enhanced performance comparing other techniques and methods such as repeat request, and repeat and migration is being used which considering low risk and risky environments parameters and metrics, reduces the round trip scheduling time of the processing resources comparing previous algorithms, hence increasing the quality of the service in different grid environments and decreases the average processing time of work. By using this new proposed algorithm, we would be able to solve some of the challenges in processing resources management and finally improve the response time for the service output which provided by the processing resources and with this new hybrid and dynamic technique overall processing time is being reduced.
Keywords :
fault tolerant computing; genetic algorithms; grid computing; scheduling; SQEFG; dynamic genetic algorithm; fault tolerance control; genetic-based algorithm; grid environment; reliable distributed resources; round trip scheduling time; scheduling algorithm; Artificial neural networks; Genetic Algorithm; Grid Computing; Resource Management; SQEFG; Scheduling;
Conference_Titel :
Communication Software and Networks (ICCSN), 2011 IEEE 3rd International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-1-61284-485-5
DOI :
10.1109/ICCSN.2011.6014946