Empirical analysis of certificate revocation lists

Abstract

Managing public key certificates revocation has long been a central issue in public key infrastructures. Though various certificate revocation mechanisms have been proposed to address this issue, little effort has been devoted to the empirical analysis of real-world certificate revocation data. In this paper, we conduct such an empirical analysis based on a large amount of data collected from VeriSign. Our study enables us to understand how long a revoked certificate lives and what the difference is in the lifetime of revoked certificates by certificate types, geographic locations, and organizations. Our study also provides a solid foundation for future research on optimal management of certificate revocation for different types of certificates requested from different organizations and located in different geographic locations. © 2008 Springer-Verlag Berlin Heidelberg.