2004 Achievments

التفاصيل

الزيارات: 1490

BEST PAPER, Best Paper Award in the 9th IEEE Symposium on Computers & Communications (ISCC), 2004Alexandria, Egypt

Title: Modeling and Analysis of a Multi-Level Caching in Distributed Database Systems

Authors: Doaa Saad El-Zanfaly, Prof. Ahmed Sharaf El-Din Ahmed, and Prof. R. Ammar (Computer Science and Engineering Dept., University of Connecticut, Ct, USA)

Abstract:

Caching frequently asked queries is an effective way to improve the performance of both centralized and distributed database systems. Intensive works have been done in this area to propose different query caching techniques and to evaluate their performance. However, most of these works were confined to caching previous query results in a single-level caching architecture. Evaluations of these works were based on simulations. In [1], we proposed a new query caching technique for caching both query results and execution plans in a multi-level caching architecture. The centralized version of this technique was evaluated and the results were reported in [2]. In this paper, we present an analytical model to evaluate the performance of the proposed technique in distributed database systems.

(Figure Above) Dr. Doaa Saad Hassan El-Zanfaly, a lecturer at the Information Systems Dept., the Faculty of Computers & Information – Helwan University.

2005 Achievments

التفاصيل

الزيارات: 1501

FIRST PLACE, the Best Helwan University’s Ph.D. dissertation in the field of Engineering Sciences & Technology, 2005

Title: Efficient Query Processing In Heterogeneous Distributed Databases

Researcher: Dr. Doaa Saad Hassan El-Zanfaly

Supervisor: Prof. Ahmed Sharaf El-Din Ahmed, Prof. Abdel-Badeeh Mohamed Salem (Computer Science Dept., the Faculty of Computers & Information, Ain Shams Univ., Egypt), and Prof. Reda Anwar Ammar ( the Head of the Computer Science Dept., University of Connecticut, USA)

Abstract:

Query caching plays a vital role in improving query processing in both centralized and distributed database systems. Most query caching techniques are confined to caching prior query results in a single level caching architecture to avoid accessing the underlying database each time a user submits the same query. In this thesis, we present a new query caching technique that allows caching a combination of both plans and results of previously issued queries in a multi-level caching architecture. The objective is to reduce the response time of distributed query processing and hence increase the system throughput.

Caching query execution plans makes the cache more independent of the database. Each time the data in a database is modified, we do not need to propagate these modifications to the cache. The plan will be reused whenever needed with the new values of its attributes. On the other hand, cached plans can be reused to execute different queries instead of being used to execute only a sole general query. This reduces the processing time by avoiding re-optimization and reconstruction of global execution plans (subplans). Furthermore, dividing the cache into levels reduces the cache search time as the most frequently used plans (subplans) are cached at the top cache level.

We developed an analytical performance model to evaluate the multi-level caching technique in centralized database systems. The technique is evaluated and the results are compared to regular query optimization and single-level caching techniques. Then we extended the model to analyze and evaluate the performance of the new technique when it is applied in distributed environments. We developed a generic analytical performance model that can be applied in both homogeneous and heterogeneous distributed database systems.

Most evaluations of query caching techniques have been carried out by simulations. Analytical models were only used to evaluate cache replacement algorithms. We developed an analytical performance model that evaluates the performance of the multi-level caching technique when it is applied in centralized system, distributed homogeneous systems, and distributed heterogeneous systems. Unlike other distributed database performance models, our model is not based on the homogeneity assumption, where all database sites and their respective workloads are identical. Our model can be applied to different distributed database architectures.

Our results show that caching query execution plans in a multi-level caching architecture significantly improves the query optimization time comparable to regular query optimizers and single level caching in both centralized and distributed systems.

(Figure Above) Dr. Doaa Saad Hassan El-Zanfaly, a lecturer at the Information Systems Dept., the Faculty of Computers & Information – Helwan University.