Chapter 5: Sustaining Manageability

Manageability is one of the core attributes that affects the total cost of ownership (TCO) of a computing environment. This is because labor costs are usually one of the largest cost category in any IT operation, particularly in the case of the Windows-based datacenter because it benefits from significantly less expensive hardware and, often, licensing costs. The manageability of the Windows environment is thus a powerful economic argument in considering migration from the mainframe.

On This Page

Mainframe Manageability Overview Mainframe Manageability Overview
Windows Server 2003 Manageability Windows Server 2003 Manageability

Mainframe Manageability Overview

Mainframe manageability has traditionally been increased by concentrating the complexity and sophistication of the system into a central location. This response is based on the belief that centralization reduced complexity, and that dumb terminals do not require much management. The rise of personal computer workstations and multitiered application architectures inevitably challenged this strategy.

The personal computer workstation requires significant management. As users became more familiar with personal computers, it was tempting to defer management of the workstation to the departmental or end-user level, and continue to focus professional management skills on the central mainframe computer.

This was not unreasonable at a time when personal workstations were often functionally equivalent to terminals in the application architecture. However, as client-server and multitiered application architectures emerged, many mainframe shops were unprepared to support the entire application and could handle only the centralized components.

The development of the Internet-based e-business application also challenged this model. This type of application is inherently multitiered and, therefore, difficult to centralize its management. Merely identifying the centralized location of these applications proved difficult. For example, one could not determine whether the DBMS, the Web server, or perhaps the procedures that ran on middle-tier servers are the central location.

Increasingly, the mainframe manager must confront the problem that managing the mainframe is not managing the whole application. The tools available for managing mainframes are not designed for managing all of the other components. This results in an inevitable proliferation of tools and many new skills to be mastered.

In response to this complexity, one option was the idea of recentralizing all application components on the mainframe as a way to recapture the simplicity and, therefore, the manageability of the old centralized architectures.

Executing all components of the modern application on the processors contained within a single mainframe configuration may be conceptually appealing, but is seldom appealing in practice. The fact that it can be done at all is predicated on the mainframe's capability to simultaneously emulate many different environments, and even different operating systems, while sharing resources equally among those diverse needs. However, each of those environments comes with its own overhead in terms of processing and unique tools and skills. Regardless of the fact that all of those might be executed within a single physical unit, the truth is that the complexity exists whether it can be physically perceived or not.

Another aspect to simplification through recentralization is consolidating the workload of multiple servers into a single mainframe partition. The wisdom of concentrating services that have a widely-dispersed user base into a small logical network footprint is debatable. The economic cases made for server consolidation are often based on two rather artificial assumptions:

  1. Windows technology servers are typically run at utilization levels on the order of 5 to 10 percent.

  2. Computing power per dollar is roughly equivalent on the mainframe and the Windows Server.

The first assumption represents a worst-case scenario of uncontrolled server proliferation, a problem which can be rectified inexpensively by consolidating Windows Servers where appropriate, instead of by porting the workload to a mainframe. The second assumption is clearly invalidated by numerous benchmarks that show that mainframe power is significantly more expensive on a per unit basis.

For more information on Server Consolidation planning and optimization considerations, refer to:

Windows Server 2003 Manageability

Each edition of Windows Server 2003 supports the indicated level of load manageability features:

  • Windows Server 2003, Standard Edition: IntelliMirror, Group Policy Results, Remote OS Installation, Remote Installation Services (RIS), WMI

  • Windows Server 2003, Enterprise Edition: All features of the Standard Edition, plus WSRM

  • Windows Server 2003, Datacenter Edition: All features of the Enterprise Edition, plus vendor-specific services and hardware.

  • Windows Server 2003, Web Edition: All features of the Enterprise Edition, except RIS and WSRM

Many aspects of system manageability are critical for the overall success of a Windows Server System production environment:

  • Patching, or applying corrective changes to, system software

  • Monitoring system health to predict and avoid problems

  • Deploying new or corrected software configurations to all servers and desktops as required

  • Employing skilled and experienced administrators

  • Maintaining mature and formalized operational procedures

Deploying and Managing Updates

All enterprise operating systems are supported by a steady flow of patches that require installation. Too often IT professionals who maintain mainframe systems are not the same people who are responsible for maintaining the Windows-based file sharing and e-mail servers, and the tribal knowledge of how to sustain a production computing environment is not shared.

Microsoft Systems Management Server (SMS) is a mechanism for deploying and managing the distribution of software updates to a large number of clients. SMS is capable of optimizing the deployment of applications and security patches by identifying system capabilities of users and locations and adjusting download and update properties accordingly. SMS provides the following functionality:

  • Inventory functions to determine how many computers have been deployed and to identify their locations and roles

  • Inventory functions to identify which software applications and software updates have been installed and which need to be installed on the deployed computers

  • Scheduling functions that allow an organization to deploy software updates outside regular working hours, or at a time that has the least impact on business operations

  • Status reporting that allows administrators to monitor the progress of installation

SMS inventory scanning programs create an inventory of applicable and installed updates for each client computer using an automated source of detection logic. The resulting data is included in the SMS inventory and a comprehensive view of the status is provided through the Web-based reporting capabilities. The manageability benefits of SMS are largest when automating the deployment of patches and updates to Windows desktop clients and departmental servers in a distributed computing environment. In an isolated replacement of a single mainframe by one or more Windows servers, a careful configuration of the Windows Update Service may be a better solution.

For more information on Microsoft Systems Management Server, refer to:

For more information on the Windows Update Services, refer to:

Monitoring System Health

Windows System Resource Manager provides resource management and enables the allocation of resources, including processor and memory resources, among multiple processes based on business priorities. An administrator sets targets for the amount of hardware resources that running applications or users (typically in a Terminal Server environment) are allowed to consume.

Windows Management Instrumentation (WMI) provides access to the management functions of local and remote systems. Administrators can directly access these management functions and create queries based on this data that will update remote systems on a selective basis. Administrators can monitor local and remote Windows Server 2003 and Windows XP systems directly, and view the resulting data from the command line, or retrieve it in XML form. WMI follows the specifications of the industry standard Common Information Model defined by the Distributed Management Task Force.

Group policy results allow administrators to see the effect of a group policy on a targeted user or computer. Included in the Group Policy Management Console, group policy results provide administrators with a powerful and flexible base-level tool to plan, monitor, and troubleshoot policies.

Deploying Software to Servers and Desktops

IntelliMirror provides administrators with high levels of control over portable and desktop systems running Windows 2000 Professional or Windows XP Professional. Administrators use IntelliMirror to define policies based on business roles, group memberships, and locations. With these policies, desktops are automatically reconfigured to meet a specific user's requirements each time that user logs on to the network, regardless of where the user logs on from.

Remote Installation Services allows Independent Hardware Vendors (IHVs), Independent Software Vendors (ISVs), and Original Equipment Manufacturers (OEMs) to create tools that help network administrators and end users perform tasks such as system BIOS updates, stand-alone system diagnosis, and virus scanning and repair. RIS provides a centralized location to integrate maintenance and troubleshooting tools that are accessible through a network boot.

Application Center reduces application management complexity by allowing administrators to quickly construct logical groupings including the contents, components and configuration of applications. It also applies any changes made to a server by updating other servers in the cluster, automates the deployment of applications from one server to another, and allows applications to achieve on-demand scalability.

For more information on Application Center, refer to:

Deploying Content

Very large Web applications present a management challenge to organizations because of the vast amount of content that must be constantly maintained, while the demand for access by users is high and unremitting. In addition, new channels for access to content appear with ever-increasing frequency.

Microsoft Content Management Server simplifies the process of creating, deploying, and maintaining content-rich Web pages. Content Management Server provides the capability to tailor Web pages to the needs of different browsing devices and effectively target and personalize the experience of individual site users. Content Management Server supports industry-standard load balancing and failover technologies, serves content in XML, and enables XML Web services customization and standards-based interoperability.

For more information on Content Management Server, refer to:

Managing Operations

Mainframe customers are accustomed to selecting and adapting multiple third-party products to help automate the management of their systems. Microsoft Operations Manager is a unified programmable and extensible management environment specific to the Windows Server environment.

MOM was created to help prevent potential IT problems and support issues that do arise. MOM is designed to help staff stay aware of the status of IT health, improve response times to any issues, and effectively share information about key issues.

For more information on Microsoft Operations Manager, refer to:

Microsoft Manageability Guidance

One of the most important features of the manageability of the Windows platform is that it comes with a significant amount of detailed prescriptive guidance. This guidance is in the form of patterns and practices contained in Solution Accelerators: lab-tested, customer-approved Microsoft best practices that are intended to be used by Microsoft Consulting Services (MCS) or Microsoft partners to help customers achieve optimal solutions. The following Microsoft Solution Accelerators that apply to manageability are available:

  • Windows Server Deployment. This document describes the efficient deployment of Microsoft Windows Server 2003 using MOM and SMS.

  • Business Desktop Deployment This document describes the efficient deployment of Microsoft Windows XP Professional, as well as Microsoft Office XP Professional or Office Professional Edition 2003.

  • Security Patch Management. This document describes the efficient deployment of security patches within organizations using SMS or Windows Update Services.

  • Service Monitoring and Control. This document describes the knowledge, tools, and services to monitor services running on the Microsoft platform using MOM.

  • Account Management. This document provides guidance for streamlining user account administration and office location changes.

  • New Application Installation. This document provides an automated, repeatable method for quickly and efficiently rolling out new application software.

Sources for Detailed Guidance

For more information on Microsoft Solutions for Management, refer to:

For an overview of all Microsoft Solutions for Management Prescriptive Guides, refer to:

For more information on Microsoft Management Solutions for Windows Server 2003, refer to: