International Research Journal of Engineering and Technology (IRJET) Volume: 04 Issue: 05 | May -2017
www.irjet.net
e-ISSN: 2395 -0056 p-ISSN: 2395-0072
A Comparative Study of Database Connection Pooling Strategy Sohel S. Shaikh1, Dr. Vinod. K. Pachghare2 1Student, 2Associate
Dept. of Computer Engineering, College of Engineering Pune, Maharashtra, India Professor, Dept. of Computer Engineering & IT, College of Engineering Pune, Maharashtra, India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - In recent times, with fast advancement in
applications developed using traditional database and database connection pools.
technology there is a faster need for reducing the dormant user time while interacting with software applications. Every application makes a use of databases to store their data for several purposes. The conventional database is a good fit for application with fewer database interactions. As the number of interactions increases the load on the application increases thus leading to rise in response time. This paper proposes a novel technique of database interaction with the advent of database Connection Pooling. It details about the proposed model and elaborates its working principle. The proposed methodology advances the efficiency of accessing database significantly.
The paper progresses as follows: Section 2 provides insight about the existing conventional database connection. Section 3 gives an overview of the Database Connection Pooling. Section 4 delivers the proposed methodology. Section 5 details the features that were set for the proposed system implementations. Section 6 describes the experimental results by comparing the different techniques. Section 7 concludes the accomplishments of the paper.
2. CONVENTIONAL TECHNIQUE
Key Words: Database Connection Pooling, persistent systems, performance optimization.
DATABASE
CONNECTION
1. INTRODUCTION In today’s world, there is enormous data available. These data have a great potential to expand ones’ information base. These data are needed to be stored and well retrieved as and when obligatory. To stock this data databases are used. Database is a set of information that is schematized so that it can be fluently retrieved, sustained and restructured. To smoothen these actions a database management system is established. Connections are developed to communicate with these database management systems. For every request a connection is formed, resources such as locks, transactional logs, etc. are assigned. These connections are luxurious to create, owing to the overhead of starting network acquaintances and adjusting database connection sessions in back-end chronicles. Database resources such as locks, memory cursors, transaction logs, statement handles and temporary tables increase with increase in the number of concurrent connection sessions. [1]
Fig 1: Conventional Database Connection At the present time, web applications have become an inevitable necessity for every individual. Such applications necessitate run time processing of requests from the users. Database connections are to be established on the fly to facilitate these huge number of requests. This implies substantial load on the servers to create a connection for every request from scratch, allocate resources and close the connections. Hence runtime creation of such connections becomes too expensive. Also, garbage collection becomes an additional overhead. Connection initialization involves time consuming processing to perform user authentication, establishing transactional contexts and establishing other aspects of session that are required for succeeding database usage. It is not only expensive to create but also expensive to maintain overtime. In addition, if the connections are left opened even after completion of tasks, it may lead to more serious issues like memory leakage leading the application to break.
In such scenarios, there is a compulsive need for an alternative mechanism called the Database Connection Pooling. Database Connection Pooling is a reserve of live database connections. The admission of Database pools facilitates faster connection of applications with the database management systems. Also, the chief feature is that the connections within a pool can be reused once its purpose is accomplished. This paper proposes a model implementing connection pools and provides the comparison among the
© 2017, IRJET
|
Impact Factor value: 5.181
|
ISO 9001:2008 Certified Journal
|
Page 230