A Proxy Service for Multi-tenant Elastic Extension Tables - Transactions on Large-Scale-Data-and Knowledge-Centered Systems XX - page 1

A Proxy Service for Multi-tenant Elastic
Extension Tables

Haitham Yaish 1,2( & ) , Madhu Goyal 1,2 , and George Feuerlicht 2,3,4

1 Centre for Quantum Computation and Intelligent Systems,
University of Technology Sydney,
P.O. Box 123, Broadway, NSW 2007, Australia
2 Faculty of Engineering and Information Technology,
University of Technology Sydney,
P.O. Box 123, Broadway, NSW 2007, Australia
{haitham.yaish,madhu.goyal-2,
george.feuerlicht}@uts.edu.au
3 Unicorn College, Prague, Czech Republic
4 University of Economics, Prague, Czech Republic

Abstract. An important challenge in the design of multi-tenant databases that
support Software as a Service (SaaS) applications is providing a platform
that manages multiple tenants ' data in single database storage. To address this
challenge, we have previously proposed a multi-tenant database schema called
Elastic Extension Tables (EET) that uses single shared database and shared
schema for all tenants. In this paper, we extend this work with a multi-tenant
database proxy service called Elastic Extension Tables Proxy Service (EETPS)
that combines tenants ' relational tables with virtual relational tables into single
database schema for each tenant. This service enables data access by calling
functions in order to avoid writing SQL queries and backend data management
code. It allows the tenants to focus on their core business and easily create SaaS,
mobile, web, and desktop software applications. We present the EETPS algo-
rithms and perform several experiments to assess their feasibility and effec-
tiveness by comparing the performance of retrieving data from traditional
physical tables, virtual tables, and integrated physical and virtual tables.

Keywords: Cloud Computing Software as a Service Multi-tenancy

Multi-tenant database Elastic Extension Tables

1 Introduction

Software as a service (SaaS) is a Cloud Computing service model that exploits econ-
omies of scale for SaaS service providers by offering single con gurable software and

computing environment for multiple tenants. It is an emerging software application
service model and a signi

cant current topic in the software industry. Also, SaaS can be
dened as a model where the service provider hosts the applications and customers
accesses these applications via the internet [ 19 ]. Con

guration is one of the main
characteristics of multi-tenant SaaS applications that allows SaaS vendors to run a single
instance of the application. This requires a multi-tenant aware design with a single