Off-campus UNL users: To download campus access dissertations, please use the following link to log into our proxy server with your NU ID and password. When you are done browsing please remember to return to this page and log out.

Non-UNL users: Please talk to your librarian about requesting this dissertation through interlibrary loan.

Optical Cloud Networking and Named Data Networking for High Speed Scientific Data Transfer

Mohammad A Alhowaidi, University of Nebraska - Lincoln

Abstract

The popularity of cloud computing has increased in recent years due to its ability to offer a large set of computing and storage services. The cloud infrastructure provides the users important facilities to run their tasks and experiments without worrying about the network setup across different platforms which satisfy their demands. The cloud infrastructure needs to accommodate a large number of demands and satisfy them in a timely manner, so the need for a smart network solution became a necessity. Relying on the current Internet infrastructure has several limitations in performance and security as we discuss in this work. Various network innovations have been proposed to improve the current Internet architecture such as Software Defined Networks (SDN) and Content-Centric Networking (CCN). These two architectures can be used by the cloud infrastructure providers to improve the network/infrastructure performance and increase their profits. The functional goal of the Internet Protocol as conceived and created in the 1970s was to enable two machines, one comprising resources and the other desiring access to those resources, to have a conversation with each other. The operating principle was to assign addresses to end-points, thereby enabling these end-points to locate and connect with one another. Since those early days, there have been fundamental changes in the way the Internet is used from the proliferation of social networking services to viewing and sharing digital content such as videos, photographs, documents, etc. Instead of providing basic connectivity, the Internet has become largely a distribution network with massive amounts of video and web content flowing from content providers to viewers. Internet users of today are demanding faster, more efficient, and more secure access to content without being concerned with where that content might be located. The main characteristics of the scientific data are the data size and high-speed retrieval. In our proposed work, we study two architectures which can be used to store and retrieve the scientific data. Firstly, we studied the Resources Allocation of Virtualized Cloud Infrastructure (RA-VCI) problem. In the RA-VCI problem, we used a multi-layer optical network as a network infrastructure to receive the user demands and move the data across the cloud. Secondly, we study the use of Named Data Networking (NDN) as a replacement for the current HTTP protocol to retrieve the scientific data software. Thirdly, we use both SDN and NDN to enhance scientific data transfer and management. SDN will enable multi-path data retrieval from on-path and off-path caches. Finally, we benefit from the centralized control of SDN to implement a distributed cache management using NDN. This cache management helps in caching the large files across the NDN routers. SDN will distribute the large files to multiple router’s caches and use its global knowledge of the topology and cache map to retrieve these files from multiple routers. Further, SDN enables the prefetching technique to improve data retrieval performance.

Subject Area

Computer Engineering|Computer science

Recommended Citation

Alhowaidi, Mohammad A, "Optical Cloud Networking and Named Data Networking for High Speed Scientific Data Transfer" (2019). ETD collection for University of Nebraska - Lincoln. AAI13898274.
https://digitalcommons.unl.edu/dissertations/AAI13898274

Share

COinS