On 4 May 2004, Microsoft announced changes to NGSCB that will integrate it closely with the forthcoming version of Windows, code-named Longhorn. This integration is designed to make NGSCB more accessible to applications without modification.

NGSCB was originally designed to create a parallel, secured, sealed and trusted partition that ran alongside any operating system. This architecture isolated NGSCB from the Windows development process, but also required modifications to applications using NGSCB features.
Microsoft now plans to simplify implementation of NGSCB within the Longhorn infrastructure. Longhorn will provide scheduling, boot, advanced driver, and application programming interface (API) capabilities for NGSCB, simplifying the NGSCB nexus. This simplification makes it easier to develop the trusted code required by the nexus. The nexus may also be able to support Windows APIs, allowing the nexus agents to be more sophisticated, yet still be secured. The nexus maintains ultimate discipline over the secured memory partition feature to be introduced in NGSCB-capable processors, and will presumably be capable of offering strong isolation capabilities to the Longhorn kernel, further improving system security and stability.
NGSCB capability will now be potentially far more accessible to applications without modification. The more sophisticated NGSCB environment will open the door for a broad range of infrastructure components that need to be implemented in a secured environment. Network security and hard-drive encryption would, for example, be good candidates for Nexus implementation, because they cannot be compromised by a successful attack on Longhorn.
Recommendations: Gartner expects Longhorn to launch with NGSCB components that provide strong authentication without passwords which will, for example, enable users to effectively secure laptop data. Gartner recommends that technology vendors and buyers continue to observe the progress of the Longhorn/NGSCB program and plan for availability in 2007.
Analytical Source: Martin Reynolds, Gartner Research
Recommended Reading and Related Research
(You may need to sign in or be a Gartner client to access all of this content.)

|