Firewall is a typical security system that extensively secures the private networks. The operation of a firewall is to analyze every packet and decide whether to accept or discard it based upon the firewall policy. This policy is specified as a set of rules. The work focuses on inter-firewall optimization over distinct administrative domain without exploiting the privacy policies. With the massive growth of Internet-based applications, the number of rules in firewalls has been increasing in a rapid rate, which degrades the network performance and throughput. To mitigate the number of rules validation for every session, a dynamic rules estimation algorithm is proposed. However, an error in a firewall either discloses secret information from its network or interrupts proper communication between its network and the Internet. The redundancy removal algorithm is used to overcome these problems by reducing the redundant rules in the firewall with multi-rule coverage. The optimization process involves semi-honest computation between the two firewalls by preserving privacy of the each party firewall policies. The algorithm used will avoid the rules overhead and increases the efficiency by optimizing the firewall.

Dynamic rule estimation Redundancy removal algorithm Firewall optimization.