Display Full Text | Download Full Text
1.96 MB PDF file

Degree

Master of Science, Ohio State University, Computer Science and Engineering, 2010.

Abstract

The recent emergence of cloud computing is making the vision of utility computing realizable, i.e., computing resources and services from a cloud can be delivered, utilized, and paid for in the same fashion as utilities like water or electricity. Current cloud service providers have taken some steps towards supporting the true pay-as-you-go or a utility-like pricing model, and current research points towards more fine-grained allocation and pricing of resources in the future. We consider streaming applications, where the data is generated by external sources. This, however, creates new resource provisioning problems. Because of the pay-as-you-go model, resource provisioning should be performed carefully, keeping resource costs to a minimum, while meeting application needs.

In the first part of our work, we focus on such environments where one needs to avoid both resource under-provisioning (leading to application slowdown) and over-provisioning (leading to unnecessary resource costs). The goal is to carefully allocate resources so that the processing rate can match the rate of data arrival. We have developed a solution that can handle unexpected data rates, including the transient rates.

In the second part of our work, we focus on streaming applications which have adaptable parameters that provide flexibility in the level of computational accuracy. We consider the problem where application processing is associated with a fixed accuracy goal. The problem then is to automatically adapt the application-specific parameters to meet the specific accuracy goal, and then dynamically converge to near optimal resource allocation required. We have developed a solution that can handle various application accuracy goals and can adapt both to the accuracy needs, and to the corresponding resource allocation required. This solution can also handle flows with unexpected changes in the data distribution characteristics and/or rates. We evaluate our approach using two streaming applications in a virtualized environment.

Subject Headings

Computer science

Keywords

distributed computing; multi-staged streaming applications; accuracy management; dynamic CPU resource allocation

Committee / Advisors

Gagan Agrawal (Advisor)
Christopher Stewart (Committee Member)

Document number: osu1274194090
Permalink: http://rave.ohiolink.edu/etdc/view?acc_num=osu1274194090

★

This ETD has been downloaded 1432 times (through March 2013)

This ETD was one of the top ten most frequent downloads of November 2011.