A efficient approach for data placement using replication dependencies in the cloud environment

نوع: Type: thesis

مقطع: Segment: masters

عنوان: Title: A efficient approach for data placement using replication dependencies in the cloud environment

ارائه دهنده: Provider: sina pad

اساتید راهنما: Supervisors: Mehdi Sakhaei Nia(Ph. D)

اساتید مشاور: Advisory Professors: Dr fereshte azadi parand

اساتید ممتحن یا داور: Examining professors or referees: Dr moharam mansoori zade

زمان و تاریخ ارائه: Time and date of presentation: 2023

مکان ارائه: Place of presentation: 27 room

چکیده: Abstract: Today, the act of repeating and placing data is considered essential for the cloud computing system, and the use of this approach is expanding day by day. It is also impossible to ignore the importance of data replication to prevent damage and waste of resources and higher access to data and storage of these data in a distributed form in the cloud environment. The correct use of frequent data has a direct effect on the way the system works, the amount of cost and the time of execution of works. If the cost of query processing is controlled at each data center, it is possible to automatically determine which copies of the data can be used for each user request and ensure that all relevant data is available. In this paper, we present a dynamic and flexible RFH method for data replication. In this research, B-cube network is used for communication between different nodes and data storage networks. The processing fee for each query is calculated per server. This cost can be calculated by dividing the number of queries in each server by the total time of query execution. With this information, it is possible to calculate the number of queries for each server or prioritize queries with lower costs and process them sooner. The aim of this research is to increase the amount of data access, reduce the cost of repetition and reduce the wastage of resources by using the common features that exist in frequent data, which will lead to optimization in the cloud environment. This research provides a complete review of the analysis of existing data placement schemes and their measures in the cloud computing literature. This method is based on traffic and according to the traffic pattern and the processing cost of each request that is created to access a data, it will decide to repeat or migrate data. In addition, it classifies data placement and evaluates designs based on capabilities and objectives. After comparing this method with the previous methods, the experimental results show that compared to the previous main methods, the number of replications and migrations at the data center level has decreased and the cost and response time of replication has decreased without reducing the number of requests, data center and nodes

فایل: ّFile: Download فایل