TY - JOUR
T1 - PACO-VMP: Parallel Ant Colony Optimization for Virtual Machine Placement
AU - Peake, Joshua
AU - Amos, Martyn
AU - Costen, Nicholas
AU - Masala, Giovanni
AU - Lloyd, Huw
N1 - Funding information:
JP is funded by the Centre for Advanced Computational Science at Manchester Metropolitan University.
PY - 2022/4/1
Y1 - 2022/4/1
N2 - The Virtual Machine Placement (VMP) problem is a challenging optimization task that involves the assignment of virtual machines to physical machines in a cloud computing environment. The placement of virtual machines can significantly affect the use of resources in a cluster, with a subsequent impact on operational costs and the environment. In this paper, we present an improved algorithm for VMP, based on Parallel Ant Colony Optimization (PACO), which makes effective use of parallelization techniques and modern processor technologies. We achieve solution qualities that are comparable with or superior to those obtained by other nature-inspired methods, with our parallel implementation obtaining a speed-up of up to 2002x over recent serial algorithms in the literature. This allows us to rapidly find high-quality solutions that are close to the theoretical minimum number of Virtual Machines.
AB - The Virtual Machine Placement (VMP) problem is a challenging optimization task that involves the assignment of virtual machines to physical machines in a cloud computing environment. The placement of virtual machines can significantly affect the use of resources in a cluster, with a subsequent impact on operational costs and the environment. In this paper, we present an improved algorithm for VMP, based on Parallel Ant Colony Optimization (PACO), which makes effective use of parallelization techniques and modern processor technologies. We achieve solution qualities that are comparable with or superior to those obtained by other nature-inspired methods, with our parallel implementation obtaining a speed-up of up to 2002x over recent serial algorithms in the literature. This allows us to rapidly find high-quality solutions that are close to the theoretical minimum number of Virtual Machines.
KW - Virtual Machine Placement
KW - Ant Colony Optimization
KW - Swarm Intelligence
KW - Parallel MAX-MIN Ant System
KW - Parallel Ant Colony Optimization
KW - Swarm intelligence
UR - http://www.scopus.com/inward/record.url?scp=85121116686&partnerID=8YFLogxK
U2 - 10.1016/j.future.2021.11.019
DO - 10.1016/j.future.2021.11.019
M3 - Article
SN - 0167-739X
VL - 129
SP - 174
EP - 186
JO - Future Generation Computer Systems
JF - Future Generation Computer Systems
ER -