Search

Objectively speaking, cloud workflow requires task assignment and virtual resource provisioning to work together in a collaborative manner for adaptive scheduling, thus to balance the interests of both supply and demand in the cloud service under the service level agreements. In this study, we present a survey on the current cloud workflow collaborative adaptive scheduling from the perspectives of resource provisioning and job scheduling, together with the existing cloud computing research, and look into the key problems to be solved and the future research.

Cloud workflow requires task assignment and virtual resource provisioning to work together in a collaborative manner for adaptive scheduling, thus to balance the interests of both supply and demand in the cloud service under the Service Level Agreements. Focusing on collaborative adaptive scheduling of cloud workflow task and virtual resources problem, we introduce fine-grained cloud computing system model and design a novel collaborative adaptive scheduling mechanism based on reinforcement learning and multi-agent society. Our results demonstrate the efficiency of the proposed scheduling scheme.

In this paper, for the problem of long task scheduling time and unbalanced system load in the task scheduling of cloud workflow. To minimize the task scheduling time and optimize load balancing as the scheduling goal, a Markov decision process model conforming to the cloud workflow environment is established. Based on this, a multi-objective optimization cloud workflow scheduling algorithm based on reinforcement learning is proposed. The algorithm combines Q_Learning features, adding a function with a weighted fitness value function in the Q_Learning reward function so that it can apply multi-objective optimization. The set of scheduling schemes is a Pareto optimal solution set, which can select the optimal scheduling scheme according to the user’s preference. Compared with other methods, this algorithm can reduce the execution time and optimize the system load. And this paper uses the real cloud workflow data to carry out the simulation experiment, and carries on the experiment through the simulation platform WorkflowSim. The result proves the effectiveness of this algorithm.

Effective resource management can not only satisfy resource requirements of user applications, but also optimize system utilization so that more applications can be supported by the cloud. By designing an obedient real cloud computing environment model and analyzing the executing progress of workflow, we proposed a novel resource provisioning scheme based on the multi-agent and reinforcement learning in this paper. Multi-agent based cloud resource monitor mechanism includes two kinds of agent: perception-agent and resource-agent. They are employing detection and statistics cloud platform system runtime performance information and workflows execute progress information. Then, we designed a cloud resource provisioning algorithm based on reinforcement learning. The algorithm based on various information obtains by agents, make resources provisioning decision. Experiment results demonstrated the effectiveness of the proposed cloud resources provisioning scheme.

To the problem of scheduling multiple DAG workflow applications with multiple priorities submitted at different times in cloud computing environment, a novel workflow scheduling algorithm based on reinforcement learning is proposed in this paper. In the workflow scheduling scheme, the number of VMs in resources pool is defined as state space; the runtime of user task is defined as immediate reward, and then interactive with cloud computing environment to obtain the optimization policy. We use real cloud workflow to test the proposed scheme. Experiment results show the proposed scheme not only can solve the fairness of scheduling multiple DAGs with the same priority level submitted at different times, but also can ensure that the execution of the DAGs with higher priorities cannot be influenced by the DAGs with lower priorities. More importantly, the proposed scheme can reasonably schedule multiple DAGs with multiple priorities and improve utilization rate of resources better.

How to fairly schedule the multiple workflow with multiple priorities submitted at different times has become an increasing concern in workflow management system (WMS). To solve the problem, a novel workflow scheduling algorithm based on reinforcement learning is proposed in this study. In our scheme, we first define some basic concepts of reinforcement learning in cloud computing, such as state space, action space and immediate reward. Then single DAG and multiple DAG cloud workflow scheduling algorithm based on reinforcement learning are designed respectively. Reinforcement learning sets up a policy to maximise the cumulative rewards in the long-term through the repetition of trial-and-error interactions in cloud computing environment. Finally, we analyse algorithm performance by using queuing theory. We use real cloud workflow to test the proposed scheme. Our results, on the one hand, demonstrate the proposed scheme can reasonably schedule multiple DAGs with multiple priorities and improve utilisation rate of resources better and, on the other hand, show optimisation object function achieves fair workflow scheduling in cloud computing environment.

Powder formulae are an indispensable part of prescription in Traditional Chinese Medicine. Powder formulae are characterized by good therapeutic efficacy and low dose used for their preparation. Analysis of the therapeutic application and material basis of pharmacological active substance in power formulae can enable the development of new powder formulae. This in turn can contribute to reduction of wastage of drug material, relief of shortage of herbal medicinal resources and sustainable development of Traditional Chinese Medicine.