Proceedings of ISP RAS

Web applications and data: achieving abstraction and scalability.

Andrey Poskonin.


Nowadays it has become clear that no data store is all-purpose. SQL-based systems have been prevalent for several decades though this approach leads to serious issues in modern high scalability architectures. Thus, a number of different data stores have been designed to fulfill the needs of large-scale Web applications. Moreover, a single application often tends to use multiple data stores for different purposes and data. This paper covers common approaches to data persistence implementation and data store abstraction and also reveals basic considerations for design of a multi-store persistence layer for scalable Web applications. The paper consists of five sections. The first section (Introduction) discusses a problem of data management in the context of modern approaches to Web-applications design and development. The second section describes the most popular solutions of this problem based on traditional SQL-based DBMSs. Some limitations of these solutions are considered. The third section of the paper presents an overview of adventures and limitations of approached based on NoSQL DBMSs. A proposed novel approach of combine use SQLbased and NoSQL systems to develop and run Web-applications is a theme of the forth section. The final section concludes the paper.


Web applications, scalability, SQL, NoSQL, data store abstraction


Proceedings of the Institute for System Programming, vol. 23, 2012, pp. 159-172.

ISSN 2220-6426 (Online), ISSN 2079-8156 (Print).

DOI: 10.15514/ISPRAS-2012-23-10

Full text of the paper in pdf (in Russian) Back to the contents of the volume