Search In this Thesis
   Search In this Thesis  
العنوان
Developing E-learning services based on cache strategy and cloud computing /
المؤلف
Hussein, Mohammed Khaleel.
هيئة الاعداد
باحث / محمد خليل حسين
مشرف / أحمد أبوالفتوح صالح
مشرف / حازم مختار مختار البكري
مناقش / أحمد عبدالخالق سلامة
الموضوع
Cloud computing. Web databases. Information system. Cloud computing. Web-based instruction. Computer-assisted instruction. File Retrieval. Information System. E-Learning Services.
تاريخ النشر
2016.
عدد الصفحات
137 p. :
اللغة
الإنجليزية
الدرجة
الدكتوراه
التخصص
Information Systems
تاريخ الإجازة
1/1/2016
مكان الإجازة
جامعة المنصورة - كلية الحاسبات والمعلومات - Information System
الفهرس
Only 14 pages are availabe for public view

from 137

from 137

Abstract

Profitable cloud file search systems have to route user queries over massive Web indexes under stretched constraints. Additionally, file search systems require to update their indexes regularly to integrate changes to the cloud. After every index change in database, the content of cache entries may become old, thus reducing the newness of served results.
The main aim of the study is to improve the overall cloud file search system performance with different cloud platforms and different technical capabilities and conditions. In this thesis, we first present a proposed framework which called Fast File Retrieval System (FFRS) that deal with these trouble in today’s caching for large-scale file search engines and the ability to handle the changes in index. Additionally, we introduce a method that uses a (TTL) Time-To-Live value and frequently used value to set cache entrance to select revives cached results by concerning refresh queries to back-end search clusters from database.
The proposed method prioritizes the entrances to be reorganized according to the number of access with the age of an entry in the cache. A proposed technique for crawling and parsing content of the files from cloud is introduced for enhancing cloud retrieval of files, improving the overall system usability for end-users of cloud search file systems.
The idea of the proposed File Crawling algorithm is to use the discriminative interest points generated by crawling sub-system as a feature for that page, and index this page in database with caching priority zero. Then, repeat these steps until the end of pages. The cache sub-system stores few files that always used by users. If new file must be used, then caching sub-system chooses the victim file according to LRU algorithm.
Experimental studies also show that the proposed Fast File Retrieval System is a fast and accurate system. We also making a comparative study between our proposed system and other systems to verify the results.