Models and Guidelines for Dimensioning Private Clouds

Abstract

It is a real challenge for cloud providers to offer cloud resources that are 'virtually unlimited and can be appropriated in any quantity at any time' in a cost-effective way. In this paper, we propose quantitative models that enable cloud providers to make an informed trade-off between cost and quality. Distinguishing between public and private cloud environments we consider infinite and finite source models respectively. In both cases either homogeneous or heterogeneous cloud resource requests are considered. These models can be applied to cloud dimensioning based on request blocking probability, an important SLA parameter. We derive a novel, insightful method that makes it possible to compute resource requirements in private clouds with heterogeneous resource requests. We show the importance of applying finite source models in the context of private clouds. We also use the proposed models to quantify the benefits of cloud federations.

Modeling and Analysis of Converged Network-Cloud Services

“…Since networking has a strong impact on end-to-end cloud service, a holistic performance evaluation is essential before planning a converged network-cloud environment. Performance evaluation on the sufficient number of computational resources necessary to meet a desired Service Level Agreement (SLA) in a cost-effective way has attracted extensive research interest [8,9]. [10] has proposed a modeling and analysis approach by exploiting network calculus theory to define a general profile that can characterize service capability of either a network or Cloud service.…”

“…[9] modeled the blocking probability of a cloud with a large number of Resource Units (RUs) and general service times given by a M/G/m/m + r queuing system with single task arrivals and a task buffer of finite capacity. [8] evaluated finite population and heterogeneous resource requests using the blocking probability as an SLA performance measure to dimension clouds. They showed that infinite source models may lead to an overestimation of the number of RUs.…”

“…Although the papers above are related to ours, none of them considered the tradeoff between gains due to statistical multiplexing and costs to support remote traffic. While [9] and [8] considered blocking of RUs, [10] considered network-cloud service capabilities. To the best of our knowledge, this paper is the first to bridge the two aspects in an integrated manner.…”

Cache based Side Channel Attack on AES in Cloud Computing Environment

“…The first reason is that using the cloud infrastructure made available by the cloud providers increases the expenditure [11] Storing data on cloud is charged based on both the size of data and time for which data is stored. Being able to set up a private cloud using open source software reduces this expenditure and the cloud can be modified to provide features according to the requirements.…”