Proxy firewalls function by acting as a true intermediary between the client and the server. Traditional stateful inspection firewalls simply analyze traffic to determine whether it should be allowed before passing it directly to the protected server. Proxy firewalls, on the other hand, actually establish connections with both the client and the server, allowing them to inject themselves into the connection stream. This provides them with the ability to perform deep application-layer inspection of traffic to identify application-level attacks, such as SQL injection exploits.
The main disadvantage to proxy-based firewalls is their cost. They are significantly more expensive than standard stateful inspection firewalls both in terms of actual financial cost and processing time used. The best way to compensate for this is to use the proxy features sparingly. If an application will not significantly benefit from proxy filtering, disable application filtering for that particular rule. This will help to squeeze maximum performance out of the firewall.
This was first published in June 2009