Sample Planning Worksheets

Identify Your High-Availability Needs

Your environment could include one or more of the following types of applications or services:

Database (Microsoft® SQL Server™ or other database application)
Groupware (Microsoft® Exchange Server or other groupware application)
Web service (Microsoft Internet Information Services or other Web service)
Windows Internet Name Service (WINS)
Dynamic Host Configuration Protocol (DHCP)
Internally-developed line-of-business applications
Third-party applications
File and print shares

In the following subsections, define the specifications for each mission-critical application or service you intend to deploy for use with Windows 2000 Server.

Application and Service Specifications

Name of application or service: _____________________________________

High-Availability Requirements for This Application or Service

Which network protocol does the application or service require?

TCP/IP __ IPX/SPX __

Note

Microsoft does not provide high-availability solutions that support IPX/SPX.

Does your clustering solution require any of the following?

Provide data backup __
Protect access to data __
Protect the data itself __
Protect against power outages __
Protect against network outages __
Manage cluster objects and configuration __
Coordinate with other instances of the Cluster service in the cluster __
Perform failover operations __
Handle event notification __
Facilitate communication among other software components __

Software compatibility issues:

Do you have a resource DLL? Yes __ No __
Is it possible for you to use a generic resource DLL? Yes __ No __
Does the resource DLL you have support:
Two-node clusters __ N-node clusters __
Does the application installation support:
Two- node clusters __ N-node clusters __
Does the application function correctly on Windows 2000?
Yes __ No __
Is the application stateless or does it maintain a client-side state?
_______________________________________________________________

Special hardware requirements:

Is the system on the cluster Hardware Compatibility List (HCL)?
Yes __ No __
Does the system support large memory? Yes __ No __
Are you installing Microsoft® Windows® 2000 Advanced Server on any Intel PAE-based computer systems that have more than 4 GB of random access memory (RAM)? Yes __ No __

If yes, you need to:
Check the HCL to ensure that the system and components are supported for large memory.
Bring the system and components into compliance.
Do a complete backup of the system.
Modify the boot.ini file to include the PAE switch.
Test the system to ensure that it is operating correctly.
Will the system use:
SCSI (two-node) __ SCSI switch (N-node) __ Fiber Channel (N-node) __
What network adapters do you have in your environment?
Ethernet __ Token Ring __ FDDI __ ATM __

Other _______________________________________________________
Are the network adapters on the HCL? Yes __ No __
Volume of data: _________________________________________________
Number of users: ________________________________________________
Hours of operation: _______________________________________________

Expected Changes in Size and Performance Requirements for this Application or Service

Seasonal or other planned peak loads: __________________________________

_________________________________________________________________

Expected rate of increase in users: _____________________________________

_________________________________________________________________

Expected rate of increase in data: ______________________________________

_________________________________________________________________

Data Backup and Disaster Recovery Plans for this Application or Service

Maximum outage tolerance: __________________________________________

_________________________________________________________________

Impact of downtime (check those that apply):

Lost sales __
Lost productivity __
Decline in customer satisfaction __
Missed contractual obligations or legal liabilities __
Loss of competitiveness __
Increased costs due to make-up time __
Other: _________________________________________________________

_______________________________________________________________

_______________________________________________________________

Measurable cost of downtime (define the cost per application and cost per service outage that exceeds the specified tolerable maximum):

_________________________________________________________________

Site disasters

Is off-site functionality required? Yes __ No __

Identify single points of failure:

Network hub __
Network router __
Power outage __
Server connection disk __
Other server hardware such as CPU or memory __
Server software __
WAN links such as routers and dedicated lines __
Dial-up connection __
Other: _________________________________________________________

_______________________________________________________________

If an application or service fails, what is the plan to ensure availability?:

_________________________________________________________________

_________________________________________________________________

If a service fails or an outage occurs, do you have:

RAID:
Level 0 (striping) __
Level 1 (mirroring) __
Level 5 (striping with parity) __
Spare SCSI or Fibre Channel controllers Yes __ No __
Replacement disks Yes __ No __
Uninterruptible Power Supply (UPS) protection for individual users
Yes __ No __
UPS protection for the network (including hubs, bridges, routers, and so on)
Yes __ No __

Use this checklist to begin your cluster backup and recovery strategy:

Map registry keys to each resource. __
Create a catalog to document each backup. __
Identify a safe location where you can place your backups. __
Create an emergency repair disk by using the Backup utility. __