Log in / Register
Home arrow Computer Science arrow Building the Infrastructure for Cloud Security
< Prev   CONTENTS   Next >

Security Management Layer

The security management layer is the top layer, where the platform trust assertion from the previous layer is requested and consumed. This security applications layer includes some classes of traditional security applications focused on event reporting and managing compliance and risk. Because the technologies and trusted compute pools involve platform integrity and trust, workload control, and policy enforcement, it makes perfect sense to have such applications aware of and enabled to detect, report, and act on the trust information available from the Intel TXT–enabled platforms.

In the context of the TCP use model, the security management tools of interest are:

• Workload/VM policy management

• SIEM/configuration management/monitoring

• GRC/compliance

These tools are critical for mainstreaming trust and elements of cloud security into any overall corporate security management systems. This is a crucial requirement, as IT managers do not want a new suite of tools for managing cloud security; they would very much rather see existing tools extended to include the new cloud and virtualized architectures as they adopt them. The primary motivation for these security management tools is to ensure that they have the visibility to platform trust and a set of control functions to management the lifecycle of the VMs/workloads. Though initially the monitoring and enforcement of trust might be periodic, over time we envision that these tools will provide continuous monitoring and enforcement of policies based on trust.

VM/Workload Policy Management

These tools provide the mechanism to specify and define the granular security requirements for the virtual machines and workloads, and to enforce these requirements during the lifetime of those virtual machines. Defining a security policy for a workload runs the gamut from the trivial, such as asserting “I want to run on trusted servers,” to the sophisticated. For an example of the latter, a policy definition could include “Run on servers with trust level X and only on servers that are in geolocation Y, and don't co-exist with Z type of workloads.” Today, there is no canon for policy definition, nor standards for tagging the workloads. Each of the policy management ISVs carries a particular language of definition and execution environments, with these definitions likely not to be portable or interoperable with other vendor offerings. As these capabilities mature, it is imperative that policy definitions and other matters of semantics become standardized and interoperable across vendors.

Policy tools also provide an interface to feed the following information to other security management tools in this layer of the stack, like the security event management and GRC tools. They provide:

• Auditable information about the policies that have been evaluated

• Evidence considered during policy evaluation

• Whitelists/manifests/known-good measurements considered for decision making

• Reports of decisions made, such as launch or deny workload creation or migration in a certain pool of compute servers

This information is provided in different formats while preserving integrity and maintaining the chain of trust. Hytrust VPA and McAfee ePO are examples of policy management tools for trusted compute pools.

Found a mistake? Please highlight the word and press Shift + Enter  
< Prev   CONTENTS   Next >
Business & Finance
Computer Science
Language & Literature
Political science