Container-based virtualization technology has been more widely used in edge computing environments recently due to its advantages of lighter resource occupation, faster startup capability, and better resource utilizat...Container-based virtualization technology has been more widely used in edge computing environments recently due to its advantages of lighter resource occupation, faster startup capability, and better resource utilization efficiency. To meet the diverse needs of tasks, it usually needs to instantiate multiple network functions in the form of containers interconnect various generated containers to build a Container Cluster(CC). Then CCs will be deployed on edge service nodes with relatively limited resources. However, the increasingly complex and timevarying nature of tasks brings great challenges to optimal placement of CC. This paper regards the charges for various resources occupied by providing services as revenue, the service efficiency and energy consumption as cost, thus formulates a Mixed Integer Programming(MIP) model to describe the optimal placement of CC on edge service nodes. Furthermore, an Actor-Critic based Deep Reinforcement Learning(DRL) incorporating Graph Convolutional Networks(GCN) framework named as RL-GCN is proposed to solve the optimization problem. The framework obtains an optimal placement strategy through self-learning according to the requirements and objectives of the placement of CC. Particularly, through the introduction of GCN, the features of the association relationship between multiple containers in CCs can be effectively extracted to improve the quality of placement.The experiment results show that under different scales of service nodes and task requests, the proposed method can obtain the improved system performance in terms of placement error ratio, time efficiency of solution output and cumulative system revenue compared with other representative baseline methods.展开更多
With the rise of online applications such as machine learning,stream processing,and interactive data-intensive applications in shared clusters,container cluster scheduling in data centers is facing new challenges.In o...With the rise of online applications such as machine learning,stream processing,and interactive data-intensive applications in shared clusters,container cluster scheduling in data centers is facing new challenges.In order to solve the problem that application performance and economic cost cannot be balanced in a container cluster deploying a hybrid application,this paper proposes a container cluster scheduling strategy based on delay decision under multi-dimensional constraints.Formal language-based application placement constraints were introduced,and a task reorder model was established based on delayed decision-making.The experiments show that this strategy improves application performance and cluster utilization.展开更多
In order to improve the survival rate of planting seedlings of Phoebe zhen-nan, the grading standard for one-year-old container seedlings of Phoebe zhennan was developed by using cluster analysis. The results showed t...In order to improve the survival rate of planting seedlings of Phoebe zhen-nan, the grading standard for one-year-old container seedlings of Phoebe zhennan was developed by using cluster analysis. The results showed that the quality of Phoebe zhennan container seedlings could be estimated from seedling height and ground diameter. The Phoebe zhennan container seedlings were divided into 3 grades： Grade 1 （seedling height ≥ 38 cm; ground diameter ≥ 0.65 cm）, Grade 2 （31.7 cm ≤ seedling height 〈 38 cm; 0.56 cm ≤ ground diameter 〈 0.65 cm） and Grade 3 （seedling height 〈 31.7 cm; ground diameter 〈 0.56 cm）.展开更多
The electronic structure of the clusters containing oxygen, the stacking fault and the complex in the transition metal Ni are calculated by the multiple-scattering Xa method. Energy levels,density of states and transf...The electronic structure of the clusters containing oxygen, the stacking fault and the complex in the transition metal Ni are calculated by the multiple-scattering Xa method. Energy levels,density of states and transfer of charge are obtained. Based on the calculation and analysis,the influences of impurity oxygen and structure defect on the electronic structure of the clusters are discussed, and it is found that the local Ni-o cluster with the interstitial oxygen is a stable atomic configuration.展开更多
文摘Container-based virtualization technology has been more widely used in edge computing environments recently due to its advantages of lighter resource occupation, faster startup capability, and better resource utilization efficiency. To meet the diverse needs of tasks, it usually needs to instantiate multiple network functions in the form of containers interconnect various generated containers to build a Container Cluster(CC). Then CCs will be deployed on edge service nodes with relatively limited resources. However, the increasingly complex and timevarying nature of tasks brings great challenges to optimal placement of CC. This paper regards the charges for various resources occupied by providing services as revenue, the service efficiency and energy consumption as cost, thus formulates a Mixed Integer Programming(MIP) model to describe the optimal placement of CC on edge service nodes. Furthermore, an Actor-Critic based Deep Reinforcement Learning(DRL) incorporating Graph Convolutional Networks(GCN) framework named as RL-GCN is proposed to solve the optimization problem. The framework obtains an optimal placement strategy through self-learning according to the requirements and objectives of the placement of CC. Particularly, through the introduction of GCN, the features of the association relationship between multiple containers in CCs can be effectively extracted to improve the quality of placement.The experiment results show that under different scales of service nodes and task requests, the proposed method can obtain the improved system performance in terms of placement error ratio, time efficiency of solution output and cumulative system revenue compared with other representative baseline methods.
基金the Natural Science Foundation of China(No.61762008)the Guangxi Natural Science Foundation Project(No.2017GXNSFAA198141)the National Key Research and Development Project of China(No.2018YFB1404404).
文摘With the rise of online applications such as machine learning,stream processing,and interactive data-intensive applications in shared clusters,container cluster scheduling in data centers is facing new challenges.In order to solve the problem that application performance and economic cost cannot be balanced in a container cluster deploying a hybrid application,this paper proposes a container cluster scheduling strategy based on delay decision under multi-dimensional constraints.Formal language-based application placement constraints were introduced,and a task reorder model was established based on delayed decision-making.The experiments show that this strategy improves application performance and cluster utilization.
基金Supported by Forestry Science and Technology Program of Hunan Province(2010-06)~~
文摘In order to improve the survival rate of planting seedlings of Phoebe zhen-nan, the grading standard for one-year-old container seedlings of Phoebe zhennan was developed by using cluster analysis. The results showed that the quality of Phoebe zhennan container seedlings could be estimated from seedling height and ground diameter. The Phoebe zhennan container seedlings were divided into 3 grades： Grade 1 （seedling height ≥ 38 cm; ground diameter ≥ 0.65 cm）, Grade 2 （31.7 cm ≤ seedling height 〈 38 cm; 0.56 cm ≤ ground diameter 〈 0.65 cm） and Grade 3 （seedling height 〈 31.7 cm; ground diameter 〈 0.56 cm）.
文摘The electronic structure of the clusters containing oxygen, the stacking fault and the complex in the transition metal Ni are calculated by the multiple-scattering Xa method. Energy levels,density of states and transfer of charge are obtained. Based on the calculation and analysis,the influences of impurity oxygen and structure defect on the electronic structure of the clusters are discussed, and it is found that the local Ni-o cluster with the interstitial oxygen is a stable atomic configuration.
