Abstract
On-demand resource provisioning and elasticity are two of the main characteristics of the cloud computing paradigm. As a result, the load on a cloud service provider (CSP) is not fixed and almost always a number of its physical resources are not used, called spare resources. As the CSPs typically don't want to be overprovisioned at any time, they procure physical resources in accordance to a pessimistic forecast of their loads and this leads to a large amount of spare resources most of the time. Some CSPs rent their spare resources with a lower price called the spot price, which varies over time with respect to the market or the internal state of the CSP. In this paper, we assume the spot price to be a function of the CSP's load. We introduce the concept of a parasite CSP, which rents spare resources from several CSPs simultaneously with spot prices and rents them to its customers with an on-demand price lower than the host CSPs' on-demand prices. We propose the overall architecture and interaction model of the parasite CSP. Mathematical analysis has been made to calculate the amount of spare resources of the host CSPs, the amount of resources that the parasite CSP can rent (its virtual capacity) as well as the probability of SLA violations. We evaluate our analysis over pricing data gathered from Amazon EC2 services. The results show that if the parasite CSP relies on several host CSPs, its virtual capacity can be considerable and the expected penalty due to SLA violation is acceptably low.
Highlights
Rapid growth of the Internet has encouraged the evolution of new technological concepts and paradigms, cloud computing being among them
We introduced the concept of parasite cloud service providers which can provide computational resources with an ondemand price less than the on-demand price of the typical CSPs
We proposed the overall architecture of the parasite CSP as well as its model of interaction to the stakeholders
Summary
Rapid growth of the Internet has encouraged the evolution of new technological concepts and paradigms, cloud computing being among them. Cloud computing gains benefits from different concepts such as distributed computing, grid computing, and parallel computing [1] and has realized many computer scientists’ dream of using computing resources as a utility As a result, these resources can be leased and released in an on-demand fashion and with a usage-based pricing model through the Internet [2]. An idea leveraged by a number of today service providers is to rent the spare resources with a price lower than the on-demand price. This price is called the spot price and varies over time regarding to the market and the internal state of the CSP. A user presents a bid to the CSP and can use the spare resources while the spot price is less than the bid
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
