The Virtual Machine Consolidation (VMC) algorithm is the core strategy of virtualization resource management software. In general, VMC efficiency dictates cloud datacenter efficiency to a great extent. However, all the current Virtual Machine (VM) consolidation strategies, including the Iterative Correlation Match Algorithm (ICMA), are not suitable for the dynamic VM consolidation of the level of physical servers in actual datacenter environments. In this paper, we propose two VM consolidation and placement strategies which are called standard Segmentation Iteration Correlation Combination (standard SICC) and Multi-level Segmentation Iteration Correlation Combination (multi-level SICC). The standard SICC is suitable for the single-size VM consolidation environment and is the cornerstone of multi-level SICC which is suitable for the multi-size VM consolidation environment. Numerical simulation results indicate that the numbers of remaining Consolidated VM (CVM), which are generated by standard SICC, are 20% less than the corresponding parameters of ICMA in the single-level VM environment with the given initial condition. The numbers of remaining CVMs of multi-level SICC are 14% less than the corresponding parameters of ICMA in the multi-level VM environment. Furthermore, the used physical servers of multi-level SICC are also 5% less than the used servers of ICMA under the given initial condition.
Changming ZHAO
University of Electronic Science and Technology of China (UESTC)
Jian LIU
the University of Science and Technology Beijing (USTB)
Jian LIU
the Nanjing University of Posts and Telecommunications (NJUPT)
Sani UMAR ABDULLAHI
the University of Science and Technology Beijing (USTB)
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
Copy
Changming ZHAO, Jian LIU, Jian LIU, Sani UMAR ABDULLAHI, "The Multi-Level SICC Algorithm Based Virtual Machine Dynamic Consolidation and FFD Algorithm" in IEICE TRANSACTIONS on Communications,
vol. E99-B, no. 5, pp. 1110-1120, May 2016, doi: 10.1587/transcom.2015EBP3291.
Abstract: The Virtual Machine Consolidation (VMC) algorithm is the core strategy of virtualization resource management software. In general, VMC efficiency dictates cloud datacenter efficiency to a great extent. However, all the current Virtual Machine (VM) consolidation strategies, including the Iterative Correlation Match Algorithm (ICMA), are not suitable for the dynamic VM consolidation of the level of physical servers in actual datacenter environments. In this paper, we propose two VM consolidation and placement strategies which are called standard Segmentation Iteration Correlation Combination (standard SICC) and Multi-level Segmentation Iteration Correlation Combination (multi-level SICC). The standard SICC is suitable for the single-size VM consolidation environment and is the cornerstone of multi-level SICC which is suitable for the multi-size VM consolidation environment. Numerical simulation results indicate that the numbers of remaining Consolidated VM (CVM), which are generated by standard SICC, are 20% less than the corresponding parameters of ICMA in the single-level VM environment with the given initial condition. The numbers of remaining CVMs of multi-level SICC are 14% less than the corresponding parameters of ICMA in the multi-level VM environment. Furthermore, the used physical servers of multi-level SICC are also 5% less than the used servers of ICMA under the given initial condition.
URL: https://global.ieice.org/en_transactions/communications/10.1587/transcom.2015EBP3291/_p
Copy
@ARTICLE{e99-b_5_1110,
author={Changming ZHAO, Jian LIU, Jian LIU, Sani UMAR ABDULLAHI, },
journal={IEICE TRANSACTIONS on Communications},
title={The Multi-Level SICC Algorithm Based Virtual Machine Dynamic Consolidation and FFD Algorithm},
year={2016},
volume={E99-B},
number={5},
pages={1110-1120},
abstract={The Virtual Machine Consolidation (VMC) algorithm is the core strategy of virtualization resource management software. In general, VMC efficiency dictates cloud datacenter efficiency to a great extent. However, all the current Virtual Machine (VM) consolidation strategies, including the Iterative Correlation Match Algorithm (ICMA), are not suitable for the dynamic VM consolidation of the level of physical servers in actual datacenter environments. In this paper, we propose two VM consolidation and placement strategies which are called standard Segmentation Iteration Correlation Combination (standard SICC) and Multi-level Segmentation Iteration Correlation Combination (multi-level SICC). The standard SICC is suitable for the single-size VM consolidation environment and is the cornerstone of multi-level SICC which is suitable for the multi-size VM consolidation environment. Numerical simulation results indicate that the numbers of remaining Consolidated VM (CVM), which are generated by standard SICC, are 20% less than the corresponding parameters of ICMA in the single-level VM environment with the given initial condition. The numbers of remaining CVMs of multi-level SICC are 14% less than the corresponding parameters of ICMA in the multi-level VM environment. Furthermore, the used physical servers of multi-level SICC are also 5% less than the used servers of ICMA under the given initial condition.},
keywords={},
doi={10.1587/transcom.2015EBP3291},
ISSN={1745-1345},
month={May},}
Copy
TY - JOUR
TI - The Multi-Level SICC Algorithm Based Virtual Machine Dynamic Consolidation and FFD Algorithm
T2 - IEICE TRANSACTIONS on Communications
SP - 1110
EP - 1120
AU - Changming ZHAO
AU - Jian LIU
AU - Jian LIU
AU - Sani UMAR ABDULLAHI
PY - 2016
DO - 10.1587/transcom.2015EBP3291
JO - IEICE TRANSACTIONS on Communications
SN - 1745-1345
VL - E99-B
IS - 5
JA - IEICE TRANSACTIONS on Communications
Y1 - May 2016
AB - The Virtual Machine Consolidation (VMC) algorithm is the core strategy of virtualization resource management software. In general, VMC efficiency dictates cloud datacenter efficiency to a great extent. However, all the current Virtual Machine (VM) consolidation strategies, including the Iterative Correlation Match Algorithm (ICMA), are not suitable for the dynamic VM consolidation of the level of physical servers in actual datacenter environments. In this paper, we propose two VM consolidation and placement strategies which are called standard Segmentation Iteration Correlation Combination (standard SICC) and Multi-level Segmentation Iteration Correlation Combination (multi-level SICC). The standard SICC is suitable for the single-size VM consolidation environment and is the cornerstone of multi-level SICC which is suitable for the multi-size VM consolidation environment. Numerical simulation results indicate that the numbers of remaining Consolidated VM (CVM), which are generated by standard SICC, are 20% less than the corresponding parameters of ICMA in the single-level VM environment with the given initial condition. The numbers of remaining CVMs of multi-level SICC are 14% less than the corresponding parameters of ICMA in the multi-level VM environment. Furthermore, the used physical servers of multi-level SICC are also 5% less than the used servers of ICMA under the given initial condition.
ER -