The multi-tier architecture is prevalently adopted by cloudapplications, such as the three-tier web application. It is highly desirable for both tenants and providers to provide virtual networks in an efficient and e...
详细信息
ISBN:
(纸本)9781467375887
The multi-tier architecture is prevalently adopted by cloudapplications, such as the three-tier web application. It is highly desirable for both tenants and providers to provide virtual networks in an efficient and elastic way, where tenant applications can automatically scale in or out with varying workloads and providers can accommodate as many requests as possible in the underlying network. However, due to potential conflicts between efficiency and elasticity, it is challenging to achieve these two goals simultaneously in abstracting tenant requirements and designing corresponding provisioning algorithms. In this paper, we propose an efficient and elastic virtual network provisioning solution called Easy Alloc, which is comprised of an elasticity-aware abstraction model and a virtual network provisioning algorithm. To accurately capture the tenant requirement and maintain the provisioning simplicity for providers, the elasticity-aware model enables two types of decoupling, i.e., Always-on VMs for normal load and on-demand VMs for dynamic scaling, and the bandwidth requirement of each VM for intra- and inter-tier communications. Then we formulate the virtual network provisioning as an overhead minimization problem, where the objective simultaneously considers the bandwidth and elasticity overhead. Due to the NP-completeness of this problem, we leverage two heuristics, slot reservation and tier iteration, to obtain an efficient algorithm. Extensive simulation results show that compared with a typical elasticity-agnostic method under a heavy load, Easy Alloc enables a 9% increase of request acceptance rate and a 16.8% improvement of the successful extension rate. To the best of our knowledge, this is the first work targeting at the elastic virtual network provisioning.
There is a growing trend for service providers (SPs) to migrate their multi-tier applications from local clusters to public cloud networks. In the cloud environment, it is a great concern of SPs to improve the availab...
详细信息
ISBN:
(纸本)9781479989386
There is a growing trend for service providers (SPs) to migrate their multi-tier applications from local clusters to public cloud networks. In the cloud environment, it is a great concern of SPs to improve the availability of their virtual networks, in addition to guarantees of their virtual resource requirements, such as VMs and bandwidth. However, meeting these requirements results in an inefficient utilization of physical resources, which goes against the operational goal of cloud providers. To address this challenge, in this paper, we propose an availability-aware virtual network embedding framework that simultaneously improves the availability of virtual networks and the resource efficiency. We first propose a new metric to quantify the availability cost of a multi-tier virtual network, and then formulate the embedding problem as a joint optimization of the aggregated bandwidth and availability costs. Due to the NP-hardness of the embedding problem, we devise a heuristic algorithm that can solve the problem in polynomial time. Extensive simulation results show that the proposed algorithm enables CP to achieve various trade-offs between resource efficiency and availability, and to gain more revenue (e.g., 16.1% under a datacenter load of 80%) than the availability-agonistic algorithm.
暂无评论