AX4 Hyper-V VPS
Login   Search
Skip Navigation Links
Server 2008
Managing Servers
Event Viewer
Placement
Clustering
About Nap
Overview
Consolidation
Dynamic IT
TS Session Broker
Scroll up
Scroll down
Skip Navigation Links>Server 2008

Server 2008


The core aim of Windows Server 2008 is to build in and deliver security, reliability and performance. With these principles in mind, Windows Server 2008 provides enhanced access controls, flexibility of use for your IT infrastructure and the ability to customize the computing experience based on the needs of your users.

Many of the duties confronting IT professionals involve controlling or managing the infrastructure of the organization and maintaining and validating proper security of resources and information. To maintain a highly efficient and secure environment, IT professionals must find, obtain, learn and use a wealth of tools and utilities. Because these tools usually come from different sources, they're not integrated into the operating environment, and therefore don't take full advantage of all the power and flexibility available in an OS. Windows Server 2008 addresses these issues by integrating many required tools and utilities into the OS, allowing for a more robust security model and a lower cost of operations. This section covers changes made to the core of the operating system.

Windows Server 2008 provides many more advances than can be covered in this class. Therefore, this class provides an overview of the features believed to be the most valuable to a typical IT audience. You can go to the Microsoft website to learn about all the features available in Windows Server 2008.

Understanding advancements in the Windows Server 2008 core

At the core of every OS is the kernel, which manages all the resources available on a system: central processing unit (CPU), memory and any input/output (I/O) devices such as a printer, mouse and monitor. Sitting between applications and the hardware, the kernel routes software calls to specific devices and returns responses to the applications. Because the kernel has full access to resources on the computer, its security is paramount to overall system protection. The kernel is also in a primary position for recognizing and communicating any alerts that occur within the system. Additionally, the kernel can route resource calls in a balanced manner so that no resource is unduly overworked while others remain idle.

Before the release of Windows Server 2008, many of these tasks had to be performed by third-party software, which can lead to system latency. With Windows Server 2008, several new features are introduced that address these problems.

Hardened kernel architecture

A hardened kernel is one that's less vulnerable to attack, smaller in size so that it requires fewer resources to operate and contains the minimal code to provide all the required functions. Windows Server 2008 has hardened its kernel through several different methods:

  • Selective installation: When installing Windows Server 2008, you have the option of installing only the features and services required for the server to perform the functions you need. This means you can install a minimum of the code for the kernel, plus any specific features you need to customize your server.

To build a print server, for example, you install the server core, which is the minimum code required for Windows Server 2008, plus any services and drivers needed to communicate with the printers on your network. You don't have to install any other services, such as those that handle email or file sharing.

  • Flexible kernel services: You can move any service that doesn't require highly secure access to resources out of the kernel itself. These services are delivered in a layer above the kernel layer. This is an additional level of protection for the kernel in that it closes several potential security gaps.

Services that need to communicate directly to the CPU, memory or hard drive remain within the kernel.

  • Kernel-level access and profile: You can require every service running at the kernel level to be segmented to a specific access level and to have a clearly defined profile. This service profile states what and how services have access to resources.

For example, if you have a service that requires access to the CPU for a calculation and then must be able to write a response to volatile memory, the service profile specifies the type of access the service has to the CPU and the type of write operation the service can perform. A service can't perform any functions outside of its stated parameters in the service profile. This closes potential attack vectors for applications with malicious intent.

Standardized error-reporting capability

As hardware and software go about their regular functions, they can sometimes run into an event, such as an error (minor, major or catastrophic) or something of interest (such as a change to a profile). These events are useful to programmers who are developing and fine-tuning applications, quality assurance personnel who need to determine what's happening behind the scenes and operations personnel who monitor the health of systems and applications and troubleshoot when errors are encountered. A common challenge is how to identify, report and understand events as they occur at the software and hardware level. Finding a standard way of reporting and understanding events has required third-party software and a great deal of recoding to make applications perform to these standards.

Windows Server 2008 offers a solution to this problem by building an event reporting mechanism into the kernel. The Windows Hardware Error Architecture (WHEA) format provides standards for hardware errors through plug-ins, a common format for all error types and easier and uniform identification of errors. This provides a simpler model for monitoring hardware. The way this works is that software developers can build their applications to conform to the WHEA standards. When an event occurs in the application or the hardware, the kernel recognizes the standard event output and sends it to a standard log format. This simplifies the process of writing event code and understanding event output.

Hot-swap and hot-replace advances

Chances are that you have become familiar with the concept of hot swapping through the use of removable hardware, such as thumb or flash drives. However, hot swapping essential system hardware has traditionally been more challenging. Windows Server 2003 introduced the ability to add memory to a live system. In that case, you could add memory to your system without turning the system off, and the new memory would be almost instantly available to running services.

In Windows 2008, this capability has been extended to allow replacing the existing memory. If you have a faulty memory stick in your server, for example, you can alert the OS that you want to swap it out for another memory component. The kernel reroutes live processes to other available memory and reroutes it to the new memory once you complete your swapping tasks. Additionally, you can hot-swap CPUs in Windows Server 2008 in much the same way.