| |||||||||||||||||||||||||||||
|
Bolting the Back Door with NAC Network Access Control (NAC) promises to improve security, but competing approaches have muddied the waters. In this tutorial, we navigate our way through NAC architectures from Cisco, Microsoft, Trusted Computing Group and the IETF.
In part 1 and part 2, we examined the business needs driving companies to rethink network access control (NAC). Instead of trusting every device on the LAN, NAC combines user identity, endpoint security state, and policies to dynamically decide who should be allowed to use which network resources, under what pre-conditions. Only healthy, compliant endpoints are permitted to reach authorized resources, while everyone else is blocked or quarantined. This concept may sound promising, but the devil is in the details. Today, NAC is an emerging market filled with divergent implementations and no universally-agreed standard. Here in part 3, we will explore four NAC network architectures:
We will also take a brief look at overlay NAC appliances, a near-term alternative for those who like the idea of NAC but don't want to invest in network upgrades. NAC architectures Although integration points have been proposed, these four architectures all overlap with each other to some degree. Each adds new NAC protocols and/or APIs to network endpoints, servers, and the access devices that connect them. On top of that NAC-enabled network will sit new client/server posture assessment programs. As described in parts 1 and 2 of this series, NAC can be applied to many scenarios, from controlled guest access to compliance auditing. Before we dive into architectures, let's lay out one common examplekeeping infected employee laptops off our LAN:
In theory, CNAC, NAP, TNC, or NEA could be used to implement this scenario. In practice, the kinds of clients, servers, and network devices that can be supported depend on architecture and product. To understand why, let's drill down...
|
|
|||||||||||||||||||||||||||
|
|
|||||||||||||||||||||||||||||