An economic and energy-aware analysis of the viability of outsourcing cluster computing to a cloud

Abstract

ElsevierAlfonso Laguna, CD.; Caballer Fernández, M.; Alvarruiz Bermejo, F.; Moltó, G. (2013) AbstractThis paper compares the total cost of ownership of a physical cluster with the cost of a virtual Cloud-based cluster. For that purpose, cost models for both the physical cluster and the cluster on the Cloud have been developed. The model for the physical cluster takes into account previous works and incorporates a more detailed study of the costs related to energy consumption and the usage of energy saving strategies. The model for the cluster on the Cloud considers pricing options offered by Amazon EC2, such as reserving instances on a long-term basis, and also considers using tools for powering nodes on and off on demand, in order to avoid the costs associated to keeping idle nodes running. Using these cost models, a comparison is made of physical clusters with Cloud clusters of a similar size and performance. The results show that Cloud clusters are an interesting option for start-ups and other organizations with a high degree of uncertainty with respect to the computational requirements, while physical clusters are still more economically viable for organizations with a high usage rate.

Self-managed cost-efficient virtual elastic clusters on hybrid Cloud infrastructures

“…This paradigm can solve those disadvantages with customizable virtual machines (VMs) that decouple the application execution from the underlying hardware and are dynamically provisioned and released [1]. Because of that, depending on the resource usage and the cost model, it might be convenient to deploy a virtual cluster instead of a physical one, as we concluded in a previous work [2]. Virtual clusters in the Cloud introduce profound benefits for many computational workloads, but specially for embarrassingly parallel jobs.…”

EC3: Elastic Cloud Computing Cluster

“…In a previous work [2] we concluded that, in some cases, it is interesting to deploy a virtual cluster instead of a physical one. A virtual cluster in an IaaS provider is able to get a competitive performance per price rate, but also gets important benefits from the Cloud provider such as reducing the administration costs (both personnel costs and maintenance of equipments), avoiding hiring or buying the physical building to host the infrastructure, avoiding the upfront investments in hardware, cooling systems, etc.…”

High Performance Scientific Computing over Hybrid Cloud Platforms

“…One of the main disadvantages of these platforms is the relatively large upfront investment together with the maintenance cost. For small and medium-sized research groups or organizations, the purchase of such an equipment might represent an important cost [58]. In addition, physical clusters cannot be easily enlarged or shrinked according to the dynamic requirements of the organization and they lack the ability to provide a tailored execution environment customized for each application to be executed, specially when incompatible libraries are required by different applications running on the same infrastructure.…”

“…Moreover, the ability to transparently migrate running jobs to be executed on other physical resources is not a trivial task, which is an important feature for effective workload balancing. Another important limitation is the large upfront investment together with the maintenance cost of such computing equipment for small and medium-sized research groups or organizations [58].…”

“…This paradigm can address those problems with customizable VMs, which decouple the execution of the application from the underlying hardware, where they are dynamically provisioned and released [127]. Thus, depending on the resource usage and cost model, it might be convenient to deploy a virtual cluster instead of a physical one, as suggested in a previous study [58]. Virtual clusters in the cloud are highly beneficial for many computational workloads, but particularly for highly parallel tasks.…”