DistributedNetworks DistributedNetworks

Physical Structure  «Prev 

Active Directory Structure

Physical Structure of Active Directory

Document your Active Directory site names and locations. Active Directory is the directory service for Windows 2000 and Windows Server 2003 family of products. It stores information about objects on the network. An Active Directory site is defined as one or more well-connected TCP/IP subnets. A well-connected TCP/IP subnet has a fast, reliable network connection.
The logical structure of your organization is represented by the following Active Directory components:
  1. Organizational units
  2. Domains
  3. Trees
  4. Forests
The physical structure of your organization is represented by the following Active Directory components:
  1. Active Directory sites (physical subnets)
  2. Domain controllers
When planning your SMS hierarchy design, consider your Active Directory logical layout (hierarchical forest arrangement and domain structure), and its physical structure (Active Directory site topology). Later, when planning your SMS deployment and configuration, become familiar with the more granular details of the logical structure, such as organizational units, because these can help determine how you organize collections, distribute software, and perform queries in SMS. Document your physical Active Directory structure and domain structure before you begin the planning phase.

Understanding Domain-level Policies

It is important to keep in mind that certain Windows security policies can be set only at the domain level. These include the settings in the Account Policies node:
  1. account policies,
  2. account lockout policies, and
  3. Kerberos policies.
This means that an Active Directory domain can have only one of these policies in effect at any given time: all users within a single domain will be bound to a single policy for elements like
  1. password length and complexity,
  2. frequency of password changes,
  3. PKI policies, and
  4. Kerberos settings.
The only exception to this is if you create a separate account policy on an Organizational Unit (OU) containing member servers. In this case, the local user accounts on machines within a given OU can have a different account policy apply to them. However, any domain accounts, even within a separate OU, will adhere to the domain account policy. If you have a significant portion of your user base that requires different policies for account passwords, lockouts, then you should consider creating a separate domain.
Because of the transitive trusts created by Windows 2000 and Windows Server 2003, managing multiple domains isn’t nearly as tedious as it was under Windows NT. However, maintaining separate domains will still add a level of complexity to your Active Directory environment; be sure when planning your AD infrastructure that you carefully consider these domain-level policies before creating an unworkable Active Directory structure.

Each domain controller belongs to a specific site

Site location of a domain controller determines its place in the replication topology of Active Directory

Domain controller's site location is also an important factor in logon authentication, directory queries, and service requests

A domain controller's site location is established during the installation of Active Directory.

When you create the first domain controller in Windows network, the Active Directory Installation Wizard creates the initial physical structure that consists of a single site called Default-First-SIte-Name (which you can rename)

A domain controller's site location remains consistent unless an administrator manually changes its site assignment