Module 2: Planning a Data Center Environment v
Microsoft Operations Framework: People, Process, and Technology
This topic provides a brief introduction to Microsoft Operations Framework
(MOF). Briefly discuss the MOF process model and team model. For more
information about Microsoft Operations Framework, refer students to the
white papers on the Student Materials compact disc.
Lab A: Planning a Data Center Environment
This lab gives students an opportunity to apply the planning considerations
that were identified throughout Module 2. Arrange students in small teams
and have them review a scenario that involves facilities, processes, people,
and technology. Each team identifies the point of failure in one or more of
these areas and then discusses how they could have prevented it.
Customization Information
Make sure that the students meet the following prerequisites before they are
taught this module. Students need to be able to:
Differentiate between Windows 2000 Datacenter Server and a data center
environment.
Discuss components of Microsoft Windows Datacenter Program.
Module 2: Planning a Data Center Environment 1
Overview
Identifying the Data Center Environment
Identifying People Considerations
Identifying Data Center Processes
Identifying Technology Considerations
Identifying Facility Considerations
Microsoft Operations Framework: People, Process, and
Technology
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
Microsoft
® Windows® 2000 Datacenter Server is one of the many components
of a data center environment. It is important to consider the entire data center
when creating a highly available and reliable environment; implementing a data
center environment is different from implementing other environments, such as
server rooms or desktop computing infrastructures.
This module identifies the components that make up the data center
environment, and discusses how each component significantly contributes to
the whole.
This module then briefly introduces Microsoft Operations Framework and
provides technical guidance for achieving mission-critical production system
reliability, availability, supportability, and manageability on Microsoft products
and technologies.
After completing this module, you will be able to:
Identify the data center environment.
Identify people considerations for managing a data center environment.
Identify process considerations for managing a data center environment.
Identify technology considerations for a data center environment.
Identify facility considerations for a data center environment.
Describe Microsoft Operations Framework concepts and how they apply to
a data center environment.
Topic Objective
To provide an overview of
the module topics and
objectives.
Lead-in
In this module, you will learn
about components that
make up the data center
environment and discuss
how each component
contributes to the
environment.
2 Module 2: Planning a Data Center Environment
Identifying the Data Center Environment
P
e
o
p
l
e
Technology
P
r
o
c
e
s
s
Facility
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
A data center is a computing facility and a set of management processes that
maintain mission-critical requirements in your organization. It is not simply a
server or group of servers providing an extremely high level of availability.
Rather, a data center must be treated as a conglomeration of components joined
together in a whole. A data center’s components do not include just technology
but also people, processes, and facilities. Even if you install the most reliable
technology, you can only obtain the highest level of reliability that the
technology has to offer if you have the right team of people correctly managing
a carefully designed process.
The components of a data center environment include:
Management, operations, and support staff, as well as a corporate culture
that manages the data center environment as a whole.
Clearly defined processes that detail security, analysis, change management
and documentation.
Technology that emphasizes quality hardware and software, standardized
configuration, and testing.
Manageable, available, and secure facilities.
Topic Objective
To identify the data center
environment.
Lead-in
A data center is a computing
facility and management
processes that maintain
mission-critical requirements
in your organization.
This slide provides an
opportunity to introduce how
a data center environment is
more than just the
technology. For a data
center environment to be
successful, the students
must also put equal
attention on people,
processes, and facilities.
Each of the items in the
bulleted list will be covered
in detail throughout the
module.
Module 2: Planning a Data Center Environment 3
Identifying People Considerations
Management, Operations, and Support Staff
Corporate Culture
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
The importance of people extends beyond those working directly with the data
center. The culture of the organization must also support a data center that
provides high availability and reliability.
Management, Operations, and Support Staff
The people who manage and operate the data center must understand the
processes and technology and must be dedicated to preserving the integrity of
the data center. Personnel must be highly trained, disciplined, and dedicated to
the special concerns of the data center, and it is important that the organization
ensures that data center staff has redundant skills.
Corporate Culture
Even if your data center has the highest quality technology, people, and
processes, the entire operation will not work unless your organization has a data
center culture. A company that has a data center culture treats the data center as
more than just a collection of individual parts under the name of data center.
It is important to manage the data center as a whole. Consider the impact of
decisions on all parts of the data center. Avoid focusing exclusively on
technical concerns by also considering impacts to the people, processes, and
facility.
The data center culture is not limited to the staff of the data center, but must be
a commitment to availability that goes all the way to the executive support
level. After a process has been accepted and engaged, an executive who
requests that something be added to the data center cannot circumvent it.
Executive commitment ensures the integrity of the processes and provides the
stability required to achieve the goals of the data center.
Topic Objective
To identify the human
resource considerations for
the data center.
Lead-in
The importance of people
extends beyond those
working directly with the
data center.
4 Module 2: Planning a Data Center Environment
Identifying Data Center Processes
Analysis of a Data Center Environment
Change Management Process
Data Center Documentation
Data Center Security
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
The most prevalent cause of unplanned downtime is that someone makes an
untested change to a server and that change causes an issue. To maintain high
availability in the data center, all changes must be strictly controlled. And the
best way to control changes is by implementing unchangeable processes that
describe every aspect of the data center. Processes need to be strictly enforced,
difficult to change, and fully justified. Just as a modification to any business
process requires justification, so does modification to the data center.
In the data center, changes must be implemented with caution. Improvements
are good, but they must be fully justified changes. For example, if you change
the list of people who are allowed entry to the facility or change the air
conditioner in the data center, there must be a business justification. If you add
or reconfigure servers in the data center, there must be a business justification.
After the data center processes have been established, everyone must adhere to
them. The people in the data center, some of whom may be responsible for the
processes, are still bound by those processes. A legitimate alteration should be
submitted according to established process, evaluated, and justified. Only then
can it be a candidate for implementation.
Topic Objective
To identify the process
considerations in the data
center.
Lead-in
The most prevalent cause of
unplanned downtime is that
someone makes an
untested change to a server
and that change causes an
issue.
Module 2: Planning a Data Center Environment 5
Analysis of a Data Center Environment
Workload
Resources
current point
in time
Capacity Planning
Performance Monitoring
System Tuning
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
The data center environment must be analyzed to ensure proactive planning.
The three tools that can help you analyze and adjust the relationships between
workload and resources are capacity planning, performance monitoring, and
system tuning. You can use these tools to analyze and then adjust the
relationships between:
Clients (workload) and resources.
Processor transactions (workload) and utilization (resources).
Capacity Planning
Capacity planning is the process of predicting if and when system and resource
saturation will occur. One of the most important aspects of capacity planning is
the ability to perform predictive analyses of the current situation and the way in
which this situation is changing. Two examples are hardware and workload
profiles. Hardware profiles allow the analyst to theoretically reconfigure a
system to maximize resource use, and workload profiles allow the analyst to
map current user activities to predict the effects of changes to the user
population.
Performance Monitoring
Performance monitoring is the process of periodically checking hardware and
software to determine the load placed on it. Performance data must be collected
at regular intervals to provide useful information. The performance data
collection process relies on the discovery of the various servers on the network
and supports a single measurement type that collects a set number of counters
for all server types. The single measurement type reduces network congestion
caused by multiple system measurements during the transmission or retrieval of
performance data. All measurements are started, stopped, and managed from
the central console.
Topic Objective
To identify the analysis
needs of the data center.
Lead-in
The data center
environment must be
analyzed to ensure
proactive planning.
6 Module 2: Planning a Data Center Environment
System Tuning
System tuning is the process of adjusting the hardware and software to achieve
maximum performance. After you identify problem areas through capacity
planning and performance monitoring, you are ready to make the appropriate
adjustments. You need to use caution when system tuning the data center.
Before making adjustments, you need to collect sufficient data about the system
to make decisions about how these changes will impact your data center.
Module 2: Planning a Data Center Environment 7
Change Management Process
Justification
Justification
Document
Document
Implement
Implement
Approval
Approval
Identify Issue
Identify Issue
Test
Test
*****************************I
LLEGAL FOR NON-TRAINER USE*****************************
Change management is a process that ensures stability no matter what changes
are made to the data center environment. Changes can include installing,
deleting, and modifying services or applications as well as adding,
reconfiguring, or removing hardware. A proper change management process
ensures that full justification is received for every change to the data center
environment.
Change Management Process
Change management tends to operate at two levels: planned changes and
emergencies. With planned changes, usually one or more computers are
changed, typically at a scheduled interval. The following steps illustrate a
change management process. This process can be modified for both planned
and emergency changes.
1. Identify the Issue
Identify the improvement or problem that needs to be addressed.
2. Provide Justification for the Change
When dealing with an application or a service, you must not only justify the
immediate item but also any related applications or services. For example, there
is justification for moving messaging services to the data center. But in
deciding which messaging platform to use, other issues arise. Planning for
messaging with Microsoft Exchange 2000 Server requires more than just the
Exchange service. Exchange also requires access to a domain controller, Simple
Mail Transfer Protocol (SMTP) services, and Network News Transfer Protocol
(NNTP) services. So the justification for adding messaging on the Exchange
platform to the data center must also include justification for the ancillary
services it requires.
Topic Objective
To identify change
management processes in a
data center.
Lead-in
Change management is a
process that ensures
stability no matter what
changes are made to the
data center environment.
8 Module 2: Planning a Data Center Environment
In this situation, not only do you introduce an application, but you also modify
services and perhaps generate a hardware requirement. As you examine every
aspect of the new application and service included in the data center, you must
also scrutinize the inclusion of any new hardware. The hardware added to the
data center must be high quality and reliable. But, above all else, all hardware
added to the data center must fulfill some fully justified need; it cannot be
extraneous. Each piece of equipment that you introduce establishes another
cause of possible failure and another need for management. In the data center,
the addition of more hardware and software introduces expense and takes up
valuable space and resources.
3. Obtain Approval for the Change
Obtain technical and budgetary approval of the change.
4. Test the Change
Implement the change on a test system to ensure stability and reliability.
5. Document the Change
Change control has two aspects: getting consensus on when to make a change
and making sure everyone knows what changes have been made. In many
situations, a bad problem is made worse when the people making changes are
unaware that others are making changes. This is why it is important that
everyone who maintains or makes changes to a server is aware of the
importance of good record-keeping and good communication. For information
about documenting changes, see the “Data Center Documentation” topic in this
module.
6. Implement the Change
Bring the change online and into production.
Be sure to research the OEM’s specific implementation of the
Microsoft Windows Datacenter Program requirements. Windows Datacenter
Program has a qualification process that requires OEMs to ensure that all
hardware components, hardware drivers, and software that works on the kernel
level work together smoothly on servers running Microsoft Windows 2000
Datacenter Server.
Importan
t
Không có nhận xét nào:
Đăng nhận xét