Comparative analysis of database performance for well-known cloud computing environments

العناوين الأخرى

التحليل المقارن لأداء قواعد البيانات في بيئات الحوسبة السحابية المعروفة

مقدم أطروحة جامعية

Ubaydat, Jasur Ahmad Mahmud

مشرف أطروحة جامعية

Kayid, Ahmad A.

أعضاء اللجنة

al-Bahadili, Husayn
Naum, Riyad Shakir

الجامعة

جامعة الشرق الأوسط

الكلية

كلية تكنولوجيا المعلومات

القسم الأكاديمي

قسم نظم المعلومات الحاسوبية

دولة الجامعة

الأردن

الدرجة العلمية

ماجستير

تاريخ الدرجة العلمية

2014

الملخص الإنجليزي

Cloud computing paradigm reduces hardware costs, software costs, and provides efficient use of services in internet environment.

Virtualization technology is the main component of cloud computing technology.

Thus, several types of virtualization techniques (hypervisors) are available to be used in cloud environments.

Cloud computing technology serves databases for private, public and hybrid clouds.

Each type of hypervisor has its own interaction with database depending on the type of query that is executed.

The main aim of this thesis is to study the performance of the hypervisors with cloud managers for each type of SQL query.

To be able to meet this aim it has been outlined the different types of cloud managers, hypervisors, and query types.

The findings of literature review have been deployed to design several experiments for this research.

Two experiments have been conducted to evaluate the response time for both CPU and memory.

The results showed that the type of hypervisors have an effect on the performance of SQL query types with regards to CPU and memory response times.

Thus, The Mean (ϻ)was taken to make the comparison between hypervisors and The Coefficient of Variation (CV) was taken to check the comparison results.

Thus, Xen hypervisor had a positive effect based on query type.

Hence, KVM hypervisor had a negative effects based on query type.

Xen hypervisor affect the performance of query types with 11%, 20%, and 3% for DDL, DML, and DQL respectively.

KVM hypervisor affect the performance of query types with 17%, 25%, and 4% for DDL, DML, and DQL respectively.

In addition, control parameter was taken which it is dataset size; to realize the experiment results.

The domain of dataset size was taken as three stages 1100, 2200, and 3300 rows respectively.

Thus, the results show negative effect after doubling and tripling the size of data set.

التخصصات الرئيسية

تكنولوجيا المعلومات وعلم الحاسوب

عدد الصفحات

89

قائمة المحتويات

• APA
• MLA
• AMA

لغة النص

الإنجليزية

نوع البيانات

رسائل جامعية

رقم السجل

BIM-694458