Energy-Efficient Task Scheduling Algorithms with Human Intelligence Based Task Shuffling and Task Relocation

Currently, more and more vendors such as Amazon, Google, IBM and Microsoft are dedicated to developing their cloud platforms for increasing large-scale data and more complex software systems. The cloud computing technique is rapidly changing the computing environment for various applications. However, a large number of cloud servers consume massive energy and produce huge pollution. The Smart2020 analysis shows that cloud-based computing data center and the telecommunication network will generate emission about 7% and 5% each year in 2002 and 2020, respectively. This paper aims to develop a new green algorithm that can help multiple CPUs in the cloud network not only complete the tasks before a deadline, but also greatly reduce the energy consumption. Our new green algorithm with human intelligence can effectively make task assignments via partial task shuffling and adjust the cloud servers´ speeds through smart task allocation under the time constraint. Sufficient simulation results indicate that the new green algorithm with intelligent strategies is effective compared with a traditional method and another new method. In the future, we will apply both ancient and modern human´s intelligent strategies to improve green optimization algorithms.