Technical Solution Brief

Published: December 17, 2004

Microsoft uses Microsoft Operations Manager (MOM) 2005 to monitor the operations of Exchange mailbox servers and messaging-related applications. The Microsoft IT group has established best practices for using the new MOM Management Pack for Exchange Server 2003.

This technical brief is intended for enterprise technical decision makers and messaging operations personnel who are considering deployment of Microsoft® Exchange Server 2003. The best practice recommendations and lessons learned in this paper are based on the Microsoft IT group's experience as an early adopter of Exchange Server 2003 and MOM.

Background

Microsoft IT ensures that the approximately 60,000 employees and 32,000 vendors and contractors in over 400 Microsoft locations around the world are able to access the corporate network 24 hours a day, seven days a week. Microsoft IT provides services ranging from end-user support and telecommunications management to server and network operations.

In addition to running the global IT service internally, Microsoft IT is also committed to testing Microsoft enterprise products in production before they are released to customers, with the goal to ensure that products will scale to meet the business challenges of other large enterprises. This process is known internally as "eating our own dog food." Microsoft IT's customer service mission extends to sharing its experiences, best practices, and lessons learned through papers such as this one. As of this writing, Microsoft IT has completed the internal deployment of Exchange Server 2003, Microsoft Windows Server™ 2003, MOM 2005, and the MOM 2005 Management Pack for Exchange Server 2003 within Microsoft.

The Microsoft IT group initiative to model best practices is known internally as the Model Enterprise initiative. This initiative began in fiscal year 2003 (FY03) with a Microsoft Operations Framework (MOF) assessment, conducted by an outside company that did interviews focusing first on change management and release management. These service management functions are conducted much more frequently in Microsoft IT than in comparably sized enterprises because of dogfooding requirements.

Like other enterprise IT organizations, Microsoft IT's operational goals focus on the right balance of availability, performance, flexibility, and cost. Unique is Microsoft IT's commitment to use Microsoft solutions wherever possible, often before they are released to the public.

To achieve these goals, the model enterprise strategy is based on:

• Maximizing the number of management tasks performed centrally (remotely from the device).

• Reducing the number of data centers and the number of infrastructure and application servers.

• Standardizing infrastructure and devices worldwide.

Microsoft IT is responsible for maintaining more than 150,000 computers worldwide. At the time of this writing, Microsoft IT supports 109 servers running Exchange Server 2003 on Windows Server 2003, Enterprise Edition, distributed over seven locations worldwide. Thirty-six of these servers are mailbox servers hosting more than 86,000 user mailboxes and 101,000 total mailboxes.

The corporate e-mail infrastructure at Microsoft, as of July 2004, is comprised of:

• Global e-mail flow of 3,000,000 messages per day on average, with an average of more than 10,000,000 Internet e-mail messages received per day.

• Twenty databases per server, with 50 gigabyte (GB) maximum database size on new clustered deployments, with 200 megabyte (MB) per user mailbox limits.

• Global Exchange mailbox server availability of 99.99 percent.

• Worldwide e-mail delivery in less than 90 seconds, 99 percent of the time.

• Backup and restore operation Service Level Agreement (SLA) of less than one hour per database.

The Microsoft IT group strategy is to deploy infrastructure servers such as messaging servers in dedicated roles for efficient administration. Table 1 shows how Exchange servers are distributed as of this writing.

Table 1. Messaging Server Role Distribution

* Front-end servers were consolidated with the deployment of Exchange Server 2003 because the technology formerly included in Mobile Information Server 2002 was added to Exchange Server 2003. To increase system availability, each Exchange Server 2003 front-end server deployment site was configured with a pair of load-balanced servers.

Monitoring is a key part of managing infrastructure servers and applications, especially in a large enterprise such as Microsoft.

IT Environment

The three main enterprise data centers for Microsoft are in Redmond, Washington; Dublin, Ireland; and Chofu, Japan. In addition, there are three regional data centers and approximately 400 worldwide business locations. The data center operations team offers three levels of service to the owners of managed servers in a Microsoft data center. Table 2 describes the service levels.

Table 2. Service Level Offerings

Full management is critical for efficient daily operations of the core infrastructure; such as file and print servers, proxy servers, remote access servers, and servers that run the Active Directory® directory service, Domain Name System (DNS), and Windows Internet Name Service (WINS). Service level two is often chosen for lab servers. Regardless of the service level chosen, each server owner is responsible for managing and maintaining servers at the application level and higher. For example, server owners are responsible for managing user rights.

At Microsoft, the server owners must use approved versions of server software and the latest updates. The server owners must also use hardware that is manufactured by approved vendors according to corporate standard specifications. This standardization ensures that the data center operations team can control costs by managing these servers centrally.

Escalation Hierarchy

When a problem with a fully managed server is identified at Microsoft, the problem is escalated as follows:

Tier 1: Helpdesk. Most mailbox server issues are detected at Tier 2. However, if the server owner or an application user identifies the problem, that owner contacts the Helpdesk. Smaller regional mailbox servers are centrally monitored, and operating systems and Exchange service items are centrally managed. Hardware issues for regional mailbox servers are handled locally. LocalHelpdesk technicians perform any needed hands-on server operations and provide first-line support to the user community in their native language.

Tier 2: Data center operations. The data center operations team uses MOM alerts to proactively monitor servers for problems, so many problems never reach the Helpdesk. However, if the server owner identifies the problem and contacts the Helpdesk, the Helpdesk then contacts the data center operations team or messaging operations teams for further action. After being alerted, the data center operations team handles the initial response, spending a defined amount of time, such as 15 minutes, on the issue using an internally developed Troubleshooting Guide (TSG). The data center operations team also initiates the incident ticket in the internally developed ticketing system at Microsoft. This system integrates the ticket tracking application with several knowledge management functions, such as the product group knowledge base, TSGs, and other internal resources. The ticketing system is used to manage the incident life cycle from detection to resolution.

TSGs are created when an issue is common and the resolution is known and easily implemented. If the data center operations team cannot resolve the issue, they escalate according to instructions in each TSG for investigation and root cause resolution. TSGs are linked to alerts in the custom ticketing application.

Note: For more information on Microsoft IT's use of MOM, see the IT Showcase white paper titled "Monitoring Enterprise Servers at Microsoft: Deployment of Microsoft Operations Manager 2000 SP1" at http://www.microsoft.com/services/microsoftservices/mom.mspx#60.

Tier 3: Infrastructure Support and Advanced Diagnostics and Debug teams. Depending on the nature of the problem, the data center operations team can contact either the Infrastructure Support team or the Advanced Diagnostics and Debug team. The Infrastructure Support team provides end-to-end measurement and management of core infrastructure services. The Advanced Diagnostics and Debug team specializes in debugging Microsoft Windows® operating system issues and communicates directly with the product development groups.

Tier 4: Engineering. The Infrastructure Support team contacts infrastructure engineering if resolving the incident involves modifying the IT architecture, hardware standards, or software standards.

Multi-Homed Infrastructure

The MOM infrastructure for monitoring Exchange Server 2003 at Microsoft is multi-homed. This allows different teams to use the data for different purposes. For example, the messaging operations team is interested in Exchange-specific events, performance data, and threshold alerts such as high remote procedure call (RPC) average latencies. The data center operations team is interested in different information: potential problems with the hardware, operating system, or disk subsystem.

Each team has its own set of rules, and configuration group. If a monitored condition occurs, an event is generated, and then MOM sends an alert, which automatically tickets the relevant team.

Hardware and operating system alerts are collected and forwarded to the data center operations team. Exchange alerts and performance data are collected and forwarded to the messaging operations group. A custom Data Transformation Services (DTS) job runs daily to archive Exchange key performance data for long-term trend analysis. For example, all sampled data is retained for seven days before archival. Events and resolved alerts are also retained for seven days. Six months of Exchange server availability data is retained in a Microsoft SQL Server™ 2000 database for long-term trend analysis. This multi-homed architecture is shown in Figure 1.

As of this writing, across 109 Microsoft IT Exchange servers, MOM averages 350 alerts per day, filtered from approximately 42,000 events and 6 million samples of performance data. These 350 alerts typically generate approximately 10 trouble tickets on issues classified as errors, critical errors, services unavailable, and security concerns.

Situation

During the planning stage, Microsoft IT identified four key business needs for early deployment of the Exchange Server 2003 Management Pack:

• Ensuring rapid realization of benefit. Monitoring and management are the largest cost components of a messaging service, so making those tasks easier can significantly reduce operating costs.

• Reducing the risks of mailbox server consolidation. During the Exchange Server 2003 deployment, Microsoft IT began a large project to consolidate mailbox servers. A major risk of consolidation is that more users might be impacted by server downtime. Accurate and timely monitoring is crucial to minimize the impact of downtime, and to reduce the time needed to restore service when an incident occurs.

  Note: As of this writing, the first Exchange server consolidation has been completed. For those interested in details about how Microsoft IT accomplished the consolidation and all of the processes and results that came from that project, see the IT Showcase white paper titled "Deploying a Worldwide Site Consolidation Solution for Exchange Server 2003 at Microsoft" at http://www.microsoft.com/services/microsoftservices/exchange.mspx#99.

• Maintaining the Microsoft Office Outlook 2003 client experience after mailbox server consolidation. Microsoft IT wanted to ensure that Outlook 2003 client availability and client performance would not degrade after mailbox server consolidation.

• Improving long-term trending and capacity planning. Thorough monitoring helps server availability in the short term and provides valuable data for improving service levels over the long term.

Solution

Exchange Server 2003 comes with a MOM 2000 Management Pack that further enables MOM to offer end-to-end monitoring of servers. The Exchange Server 2003 Management Pack provides approximately 200 new out-of-the-box rules and a built-in Microsoft Knowledge Base to monitor and preemptively send alerts about the performance, availability, and security of Exchange Server 2003. Better information facilitates rapid failure detection, reduces time to resolution, and reduces management complexity. The Exchange 2003 SP1 version of the MOM 2000 Management Pack includes support for monitoring mobility services, such as Outlook Web Access, Exchange ActiveSync® (EAS), and Outlook Mobile Access. It also offers the Exchange Management Pack Configuration Wizard to make deployment of the management pack faster and easier for customers. The Exchange 2003 Management Pack for MOM 2005 adds a state view that shows health-at-a-glance and a topology view that shows a graphical representation of the environment. It also enables Microsoft IT to monitor the Outlook 2003 client experience in addition to server performance.

Note: To use either MOM 2000 or MOM 2005 to monitor the messaging experience from the server all the way to the client, MOM requires that both Exchange Server 2003 and Outlook 2003 be used.

The MOM 2000 Management Pack for Exchange Server 2003 was updated by Microsoft and made available as a download. To download this management pack, see Exchange Server Management Pack for Microsoft Operations Manager at http://www.microsoft.com/downloads/details.aspx?FamilyId=56D036BF-8DD3-4993-
BF07-07F99F1D5CC4&displaylang=en.

The recent release of MOM 2005 includes updates to the management packs for both Exchange Server 2003 and Exchange 2000 Server. To download the latest version of the Exchange Management Pack for use with MOM 2005, see Exchange Server Management Pack for MOM 2005 at http://www.microsoft.com/downloads/details.aspx?FamilyId=EFDEA5C6-6835-47CB-AF4F-43F3A3E30279&displaylang=en.

Out-of-the-Box Rules

One useful out-of-the-box rule is an Exchange transaction log aging script called Exchange 2003 Verify Log Files are Being Truncated. This rule detects log files that are more than 24 hours old, indicating that a server did not successfully complete its daily backup. The rule sends an alert to the messaging operations group, which can then direct the IT staff in a regional server location to investigate and backup the server if necessary.

Another useful rule for Microsoft IT is Check Local Free Disk Space, which now supports clustered volume mount point disks. Initial support for volume mount points was offered in Windows 2000 Server. Volume mount point support within Windows Server clusters was made available in Windows Server 2003 and was the key to allowing Microsoft IT to scale the number of virtual servers per cluster in the large (16,000 mailbox) clustered servers.

Rules in the management pack also monitor for disk latencies on the system. If a disk subsystem is unable to efficiently process the input/output (I/O) required for the system, an alert occurs based on the disk latency threshold.

Custom Rules

A standard MOM processing rule sends an alert if a transaction log is older than 24 hours, indicating a possible backup failure. However, to support high availability and business continuance requirements, Microsoft IT operators run a custom processing rule that checks application event logs each evening between 20:00 and 20:10 Pacific Time (when backups are scheduled) for missing events that indicate scheduled backups have started. If the event is not present, that means backups may not have started, and operators are immediately alerted. Microsoft IT also has a custom consolidation rule to detect a successful number of backups over a given period of time. If the number is lower than expected, an alert is issued to the operator.

Microsoft IT uses custom rules with MOM 2005 for monitoring features of Exchange not specifically covered by the management pack. For example, the Exchange Internet e-mail gateway servers are monitored to check for the expected quantity of e-mail message flow. Too much e-mail message flow may indicate an e-mail denial of service (DoS) or virus attack is underway, while too little e-mail message flow may indicate a gateway server is malfunctioning or is not performing as expected and should be investigated. Other custom rules monitor the state of the real-time block list, sender and recipient filtering, and more.

Note: For more information about how Microsoft IT manages spam e-mail messages, see the IT Showcase white paper "Messaging Hygiene at Microsoft" at http://www.microsoft.com/services/microsoftservices/exchange.mspx#106.

Custom Reports

Custom reports supplement management pack out-of-the-box reports. For example, Microsoft IT monitors trended metrics from a long-term trending database that contains such data as server availability, white-space calculations, and total number of messages sent and received. Figure 2 is an example of a custom report that Microsoft IT derives from a database query.

The report in this example shows three performance counters that Microsoft IT captures for all internal mailbox servers at Microsoft:

• Messages Sent Total. Total messages sent through one of the transport protocols, SMTP, since startup.

• Messages Delivered. Messages successfully delivered to the information store.

• Messages Received Total. Messages received by the server or store, but not yet delivered.

The Microsoft Office Excel 2003 pivot table enables Microsoft IT to view how many messages were sent, delivered, and received by date, fiscal week, and fiscal month, as well as by individual server. The example shows messages sent, delivered, and received for the fiscal week of October 30 to November 5, 2004.

Benefits

Monitoring is a key part of daily Exchange operations. Microsoft IT uses MOM to manage the entire internal IT infrastructure. MOM 2005 and the Management Pack for Exchange Server 2003 provide:

• Easy rules customization. The management pack rules provide thorough monitoring out-of-the-box. You can easily modify these rules into custom rules tailored to server roles, or for troubleshooting or long-term trending. The degree of customization depends on the scale and the traffic patterns of your infrastructure. For example, you can use e-mail queue views to monitor queue fluctuations over an extended period of time, and set thresholds that are slightly over these fluctuations. In addition to improved service efficiency, this practice can help provide an early alert to possible security issues when the e-mail flow is abnormal.

• Improved budget forecasting. When trying to establish a budget for a project that involves altering the computer infrastructure, an organization must estimate how much hardware will be needed. Using MOM to monitor pilot deployment performance can improve your ability to purchase the correct amount of hardware and avoid extra costs. For example, the new management pack rules helped properly size mailbox servers during the Microsoft IT upgrade from Exchange 2000 Server to Exchange Server 2003.

• Enforced configuration management. The management pack has predefined configuration checks (including registry key settings, log file settings, message tracking logs, and more) and provides the ability to add custom configuration. Configuration changes can be made quickly. For example, when Microsoft IT changes the threshold for free disk space, the configuration change is committed worldwide on all targeted Exchange servers within approximately 5 minutes.

• Flexible reporting model. It is easy to write Web pages to build custom reports using a Microsoft SQL Server database. Microsoft IT can easily capture data for trending and reporting such as:

  • E-mail flow analysis.

  • Database size and white space. (White space is the amount of unused space in your database.)

  • Server configuration.

Microsoft IT uses the data it collects from MOM about its Exchange infrastructure for regular analysis in the following areas:

• Baseline analysis

• Preventative measures

• Conditions or processes outside of normal operating conditions

• Operational effectiveness evaluations

• Work windows

• Configuration management

• Security update management

• Operational metrics

Best Practices for Monitoring Exchange

The following sections describe tips for using the MOM Management Pack to monitor Exchange Server 2003 services based on Microsoft IT experience.

E-Mail

The threshold definitions provided in the management pack are associated with servers of all sizes. Because Microsoft IT operates servers in single service roles, the level of e-mail flow is dramatically different for each server role. You should treat SMTP gateway servers—which may handle thousands of e-mail messages every hour—as a separate entity for queue management. Adjust the threshold rules as appropriate for your business.

If you plan to modify the out-of-the-box management pack rules, you should not simply change the default rules. For users of MOM 2000, you should first disable the rule, copy the rule to a custom processing rule group, and then adjust the threshold definition in the custom rule group. This way, you still have the original rules intact, and you can manage updates more efficiently. Users of MOM 2005 can, as an alternative, use overrides to modify the default rules settings. The overall goal of either approach is to preserve the standard management pack as much as possible in its original state so future upgrades of the management pack can be easily made with as little administration or modification as possible.

In addition, you should establish long-term performance trend characteristics for your servers (for example, processor utilization, disk I/O, and RPC latencies), and then set thresholds that define an abnormal condition in your environment. After you adjust thresholds, be prepared to follow up and investigate changes in your environment. For example, if processor utilization is normally 40 percent and spikes to 80 percent for a sustained period, a problem may exist.

Large Mailbox Servers

Alerts on Exchange RPC counters—such as RPC average latency, and RPC requests—indicate situations that negatively affect the user experience. The cause of the problem is likely a network, CPU, disk, or memory bottleneck. In Microsoft IT experience, the bottleneck is usually the disk subsystem.

You should monitor latency on individual disks by examining logical disk counters, which includes the total number of disk transfers occurring on data devices per second and the level of I/O activity for any drive that supports a database. Determine how efficiently read and write requests associated with a disk are being processed by ensuring that the requests occur within a specified time period. For example, based on the long-term study of disk I/O operations on mailbox servers, Microsoft IT requires average read latencies of fewer than 15 milliseconds (ms) and average write latencies of fewer than 10 ms. Each organization should determine its own resource thresholds based on its unique environment.

You should disable event log replication within a clustered configuration in which the number of nodes determines the number of events and alerts. For example, in a two-node cluster, every time an event with a corresponding alert processing rule is generated, two alerts are sent to the console. Disabling event log replication helps reduce alert chatter. For more information on event log replication, see Microsoft Knowledge Base article 224969 "Configure Event Log Replication in Windows 2000 Cluster Servers" at http://support.microsoft.com/kb/224969.

Antivirus

Microsoft IT antivirus servers are gateway servers at the network perimeter. These servers filter e-mail messages before they reach the store. Use MOM to check the availability of your antivirus service. For example, by means of a Configuration Wizard provided within the default Exchange Management Pack, engineers in Microsoft IT simply select the antivirus service to be monitored. The Configuration Wizard, which greatly simplifies the process of configuring MOM to monitor any Exchange service, configures the antivirus service on every antivirus server to be checked every few minutes to make sure that the service is running. If the service is not running, MOM issues an alert.

Adjust threshold definition rules by examining the number of viruses being removed from messages. For example, every few minutes, Microsoft IT samples the counter of the number of viruses being removed. If the virus queue exceeds 100 over five sampling periods, an alert occurs. Measuring rules help Microsoft IT understand the trend in the environment for infected incoming messages, outgoing messages, outgoing viruses, and the total number of infected messages being processed. This helps serve as an early alerting mechanism in the case of certain types of DoS attacks.

Note: For more information about how Microsoft IT manages malware in e-mail messages, see the IT Showcase white paper "Messaging Hygiene at Microsoft" at http://www.microsoft.com/services/microsoftservices/exchange.mspx#106.

Client

This type of monitoring is new in the management pack and useful for SLA trending or troubleshooting. Outlook 2003 clients by default report status and connection state to Exchange 2003 servers, including failed RPC requests and RPC latency. Exchange Server 2003 aggregates this data as application log events, and Performance Monitor data. This data can be viewed through Windows Management Instrumentation (WMI) and the Exchange System Manager. You can control which client performance data is collected based on the reported client bandwidth by the Exchange server registry value ClientMonitoringMaxLowBandwidth (DWORD). This data helped Microsoft IT confirm that client experience did not degrade during a large mailbox server and location consolidation project.

MOM Deployment

Install MOM reporting on a server that is separate from the MOM database server and the Agent Manager server. This configuration causes report generation times to increase due to network latency, but it prevents adverse effects on the performance of the MOM database and DCAM servers. To avoid potential performance issues, do not run MOM reporting against your production MOM database during peak hours.

Maintain at least 40 percent free space in the MOM operational database to allow indexing and database jobs to complete successfully. Collect performance data on the MOM agents on a staggered basis to avoid overloading the MOM database with performance data.

Create two additional databases—a holding database for long-term archiving and a data warehouse for historical trending. You can then maintain important historical MOM data while keeping the MOM database at optimum levels for daily operation.

Groom your database aggressively but wisely. For example, because of the large volume of data generated from aggressive monitoring, Microsoft IT runs grooming jobs at night on an hourly basis. Configure the frequency of grooming jobs according to the size of your organization. You must have enough time between grooming jobs to allow for growth and event storms, but you must also maintain the minimum of 40 percent free disk space in the MOM database. If necessary, run grooming jobs at specific times (for example, 10 minutes after the hour and after the half hour) to avoid conflict with other processes. Set an alert rule with automated notification for grooming job failures. Use the grooming history table to obtain useful statistics for monitoring and tuning grooming.

Note: For more information about how Microsoft IT deployed MOM 2005, see the IT Showcase white paper "Deploying MOM 2005 at Microsoft" at http://www.microsoft.com/services/microsoftservices/mom.mspx#109.

Lessons Learned

The lessons that Microsoft IT learned from early deployment of the Exchange Server 2003 Management Pack at Microsoft include:

• Use MOM to monitor Exchange Server 2003 mailbox servers, antivirus gateway servers, and storage area network (SAN) storage. The metrics that MOM collects facilitated Exchange upgrade and server consolidation, because they helped to appropriately time and size server upgrades and user moves. The worldwide management of all mailbox server configuration changes is easier from a single console.

• Use MOM to validate planning assumptions for the migration from Exchange 2000 Server to Exchange Server 2003. In the past, the only way for an IT department to know if planning assumptions were inaccurate was when problems surfaced. With better performance monitoring in MOM, you can run a pilot deployment, measure the key performance indicators, and then adjust production designs and deployment schedules based on accurate information.

• Create custom MOM management packs when deploying features and services not found in the default Exchange Management Pack. Exchange Server 2003 offers a collection of new services and features, such as real time block lists, sender filtering, and recipient filtering, which may not be monitored in the default MOM Management Pack. Other components found in a typical messaging environment, such as antivirus, will also not be specifically found in the default Exchange Management Pack. MOM enables users to quickly and easily create custom management packs for monitoring these new features if they are implemented in your environment. The deployment plan for these features should always include plans for setting up monitoring through a custom MOM management pack.

• Take advantage of MOF. The MOM database contains information about current and past performance. MOF best practices recommend using this information to make more accurate predictions about future needs and improve MOF service management functions, such as capacity planning or incident management.

• Record a baseline of your entire Exchange environment with MOM, so you can better understand how to set meaningful alerts and what the alerts tell you about a sudden change in your operational environment. Knowing when an event is outside of the expected norm for your environment is critical in being able to respond appropriately when an alert appears. Keep those historical baselines for comparisons at a later date.

Conclusion

The early deployment of Exchange Server 2003, along with the MOM management pack deployment and adoption of MOF best practices in support of the Model Enterprise initiative have together substantially improved messaging management at Microsoft. Better, faster, more accurate information facilitates rapid detection and resolution of problems in the short term. Over the long term, the Management Pack for Exchange Server 2003 is providing the following solutions to Microsoft IT business needs:

• Ensuring rapid realization of benefit. The management pack makes monitoring and managing Exchange servers easier as soon as it is deployed.

• Reducing the risk of mailbox server consolidation. The management pack provides accurate and timely monitoring to minimize downtime. During the Exchange Server 2003 deployment, Microsoft IT began a large project to consolidate mailbox servers. A major risk of consolidation is that more users on a server mean more potential for server downtime costs. Better monitoring information about the end-to-end service helps manage this risk.

• Verifying Outlook 2003 client experience after mailbox server consolidation. With Exchange Server 2003 and Outlook 2003, the client experience—measured by RPC errors and round trip latencies—provides detailed information about communication between the client and Exchange. By using MOM to monitor these client perspective quantities, Microsoft IT was able to determine client availability and client performance after the consolidation of mailbox servers.

• Improving long-term trending and capacity planning. Thorough monitoring helps server availability in the short term and provides valuable data for improving the infrastructure over the long term.

• Improving long-term data trending using custom MOM reports. You can define and create your own custom reports to specifically hone in on the data that is most important for managing your environment.

For More Information

To view additional IT Showcase material please visit
http://www.microsoft.com/itshowcase

Exchange Server TechCenter
http://www.microsoft.com/technet/prodtechnol/exchange/default.mspx

Exchange Server 2003 Technical Documentation Library
http://www.microsoft.com/technet/prodtechnol/exchange/2003/library/default.mspx

Troubleshooting Exchange Server 2003 Performance white paper
http://www.microsoft.com/technet/prodtechnol/exchange/2003/library/e2k3perf.mspx

MOF Process Model for Operations white paper
http://www.microsoft.com/technet/itsolutions/cits/mo/mof/mofpm.mspx

Monitoring Enterprise Servers at Microsoft white paper
http://www.microsoft.com/services/microsoftservices/mom.mspx#60

Knowledge Base article "Configure Event Log Replication in Windows 2000 Cluster Servers"
http://support.microsoft.com/kb/224969

For more information about Microsoft products or services, call the Microsoft Sales Information Center at (800) 426-9400. In Canada, call the Microsoft Canada information Centre at (800) 563-9048. Outside the 50 United States and Canada, please contact your local Microsoft subsidiary. To access information via the World Wide Web, go to:

http://www.microsoft.com

http://www.microsoft.com/itshowcase

http://www.microsoft.com/technet/itshowcase

For any questions, comments, or suggestions on this document, or to obtain additional information about Microsoft IT Showcase, please send an e-mail message to:

showcase@microsoft.com