TY - GEN
T1 - An VM scheduling strategy based on hierarchy and load for openstack
AU - Xu, Zhigang
AU - Xiao, Limin
AU - Zhan, Weidian
AU - Yue, Xichun
AU - Ruan, Li
AU - Liu, Rui
N1 - Publisher Copyright:
© 2016 IEEE.
PY - 2017/7/13
Y1 - 2017/7/13
N2 - In the cloud computing environment, one of the most important module is the Scheduler. As the most popular open-source cloud platform, OpenStack provides us with a massive amount of scheduling strategies. But there is no one considering of the hierarchies of the VMs and hosts. We will guarantee the security of VM through these hierarchies. Although OpenStack is abundant in scheduling strategies, none of them is based on the network load of the host. This paper proposes a scheduling strategy based on the hierarchies and load. We will define the service levels and security levels for the VMs and hosts, then filter out the hosts which do not have the corresponding levels with the VMs. Each of the remaining hosts will get a weight value according to their overall load: such as CPU, memory, disk, network. In the end, the host with the highest weight value will be selected to create a VM. We build a prototype system on OpenStack to demonstrate our design and test the result of our solution. According to the experiments, VM can be created on an appropriate host.
AB - In the cloud computing environment, one of the most important module is the Scheduler. As the most popular open-source cloud platform, OpenStack provides us with a massive amount of scheduling strategies. But there is no one considering of the hierarchies of the VMs and hosts. We will guarantee the security of VM through these hierarchies. Although OpenStack is abundant in scheduling strategies, none of them is based on the network load of the host. This paper proposes a scheduling strategy based on the hierarchies and load. We will define the service levels and security levels for the VMs and hosts, then filter out the hosts which do not have the corresponding levels with the VMs. Each of the remaining hosts will get a weight value according to their overall load: such as CPU, memory, disk, network. In the end, the host with the highest weight value will be selected to create a VM. We build a prototype system on OpenStack to demonstrate our design and test the result of our solution. According to the experiments, VM can be created on an appropriate host.
KW - OpenStack
KW - hierarchy
KW - load
KW - schedule
UR - https://www.scopus.com/pages/publications/85027461583
U2 - 10.1109/CCBD.2016.022
DO - 10.1109/CCBD.2016.022
M3 - 会议稿件
AN - SCOPUS:85027461583
T3 - Proceedings - 2016 7th International Conference on Cloud Computing and Big Data, CCBD 2016
SP - 58
EP - 63
BT - Proceedings - 2016 7th International Conference on Cloud Computing and Big Data, CCBD 2016
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 7th International Conference on Cloud Computing and Big Data, CCBD 2016
Y2 - 16 November 2016 through 18 November 2016
ER -