New Approach for Single Sign-on Improvement using Load Distribution Method

Abstract

Single sign-on (SSO) is an essential tool for company and organizations to secure access the website or application by using username and password. It is used instead of local database, Lightweight Directory Access Protocol (LDAP) and Active Directory (AD). The number of new sessions on single sign-on is considered on the large organization. The large organizations have many websites and applications that need the secure access. The new sessions and concurrent sessions support are one of the key factors for determining the efficiency of single sign-on. In general, single sign-on can support only hundred scales of the new sessions and thousand scales of the concurrent sessions. In this research, we propose in-memory data store method for single sign-on session storing and load distribution method to improve single sign-on performance. The result showed that in-memory data store and load distribution methods could increase the new sessions support and concurrent sessions support up to ten thousand scales and hundred thousand scales, respectively.