الفهرس 
OverviewOnly 14 pages are availabe for public view
 |  | from | 116 |  |  |
| | | from | 116 | | |
Abstract
Cloud computing is becoming increasingly popular as it enables users to save both development and deployment time. It also reduces the operational costs of using and maintaining the systems. Moreover, it allows the use of any resources with elasticity instead of predicting workload which may be not accurate. There are many technologies that can be merged with Cloud computing to gain more benefits. One of these technologies is data warehousing, which can benefit from this trend when it’s used to save large amounts of data with unpredictable sizes and if used in distributed environments. Large amounts of data are generated daily, according to the wide usage of social media websites, scientific data and all live matters; these amounts of data should be utilized.
The Big data storage is one of the most critical tasks as it will affect the data access, analysis, retrieve and also query answering process so it can help decision makers but the traditional database concepts are insufficient. Cloud computing is very essential in big data storage process, as big data will utilize the cloud features as the elasticity. One of the most beneficial algorithms to deal with the big data is the MapReduce algorithm.
In this thesis a system is presented which consists of: a cloud based view allocation algorithm which is presented to enhance the performance of the data warehousing system over a Peer-to-Peer architecture. The proposed approach improves the allocation of the materialized views on cloud peers. It also reduces the cost of the dematerialization process and furthermore. the proposed algorithm saves the transfer cost by distributing the free space based on the required space to store the views and on the placement technique. Furthermore, the proposed algorithm saves the transfer cost by distributing the free space on the peers based on the required space to store the views.
We also proposed an algorithm for big data allocation based on a peer-to-peer cloud architecture integrates the OLAP and MapReduce over cloud (considering workload balance) in order to enhance the performance of query processing over big data, this data is stored in a form of cubes segmented into chunks and query answering technique will run based on MapReduce approach. This process is done using the proposed allocation approach to save resources and query processing times. The proposed system achieves enhancements as time saving in query processing, network transfer cost and in resources usage and utilization.
The thesis is organized as follows:
Chapter 1 (Introduction): presents an introduction to the field of cloud computing and big data. Also this chapter presents the problems on data storing and allocation on peer-to-peer cloud computing and motivation behind the proposed work. In addition the objectives of the proposed work and our contributions are listed in this chapter.
Chapter 2 (Background): discusses cloud computing main concepts and provides an overview of big data allocation and processing. Also it presents an overview on the viewpoints of cloud computing. Data warehouse and big data cubes.
Chapter 3 (Related Works): presents a survey on the related works to our research.
Chapter 4 (An Enhanced Cloud based View Materialization Approach for Peer-to-Peer Architecture): describes the proposed cloud allocation approach applied over the materialized views of the data warehouse. Also the results and analysis of the proposed approach is presented.
Chapter 5 (A Peer-to-Peer Architecture for Cloud Based Data Cubes Allocation): presents the architecture of the big data cubes allocation over the peer-to-peer cloud computing. Also it shows the experimental evaluation, and the results and analysis of the proposed algorithms.
Chapter 6 (An Enhanced Peer-to-Peer Cloud based Chunks Allocation for Big Data Cubes): introduces big data allocation approach over the cloud but after segmented into data chunks instead of data cubes and the experimental evaluation, and the results and analysis of the proposed algorithms.
Chapter 7 (The Whole Architecture Evaluation): the system evaluation includes evaluation of each algorithm and the comparisons and charts of the results.
Chapter 8 (Conclusion and Future Work): gives a conclusion of the proposed work. A summary of the proposed approaches and algorithms is presented in this chapter. Also, this chapter presents the directions of our future research.