1. Make sure the carrier has a variety of network options.
A virtual private network must work with whatever network connection your organization currently uses. This may be problematic because telecommunications service providers typically support only a narrow range of connectivity options. Most support dial-in access, many support low-speed dedicated connections, and a growing number have wide-scale support for broadband links. Before moving ahead with a deployment, you need to make sure your carrier can support your network configuration.
2. Keep the new network's overhead in check.
By adding more traffic and overhead (from the security functions) to your network, VPNs can create bottlenecks. When adding new VPN capabilities, you need to have strong network performance- monitoring tools, so you can see how well your network is functioning and identify places where upgrades may be needed.
3. Quarantine rogue devices.
Secure Sockets Layer (SSL) VPNs provide users with flexibility when accessing the agency network. In some cases, they may work from a kiosk, a home computer or a customer's system — any of which could be infected with a virus. Set up a special network segment for those users so you can identify problem data transfers and keep them from disrupting your network.
4. Know SSL's limitations.
Data transmitted using SSL is only secure on its way between the browser and the Web server. If that data is subsequently moved to another location, security may be compromised. If, for example, a Web-hosting company collects data via an SSL site, but then forwards it to you via e-mail, it may not be secure on the second part of the journey. So make sure information is locked down at all points along a connection.