Lab 3.1: Planning RAID Configurations

Lab Objectives

After completing this lab, you’ll be able to

• Design various RAID configurations to support fault-tolerant data storage

About This Lab

In this lab you’ll design several RAID configurations. The configurations will provide fault tolerance for data storage systems in various types of Windows 2000 Server computers, including file servers, domain controllers, and database servers. Each exercise within the lab focuses on one type of server configuration.

Before You Begin

Before you begin this lab, you must be able to

• Configure RAID-1 and RAID-5 storage configurations in Windows 2000 Server

Scenario: Fault-Tolerant Storage for Lucerne Publishing

Lucerne Publishing is updating its network to implement fault tolerance in the data storage systems on various Windows 2000 Server computers. The company is upgrading four types of servers: two types of file servers, servers configured as domain controllers and DHCP and DNS servers, and database servers. You’ll be using the software implementation of RAID that’s available in Windows 2000 Server to provide fault tolerance. The upgrades will consist of implementing RAID-1, RAID-5, or both, on the four types of servers. You must design the RAID configuration for each type of server.

Exercise 1: Planning a File Server Configuration

Lucerne Publishing maintains a number of file servers on which users store various types of information. The file servers represent the greatest number of services in the network. You want to set up a fault-tolerant storage system on these servers, but you want to do it as inexpensively as possible. You plan to create two logical partitions: one for the operating system (partition C:) and one for the user data (partition D:). Partition C: should be 10 GB, and partition D: should be 30 GB. For this configuration, you’ll use RAID-1 on two 10-GB hard disks and RAID-5 on four 10-GB hard disks.

On the diagram shown in Figure 3.5, you’ll need to label the logical partitions, the logical disks, and the physical disks. You’ll also be including the size of each partition and disk.

Figure 3.5 - RAID configuration for a file server

The diagram should provide a conceptual overview of how RAID will be implemented in the file servers. For example, the box in the upper-left corner of the diagram might read, "10-GB logical partition C:—Operating system," and the box beneath that might read, "RAID-1—10-GB logical disk."

1. How should you label the rest of the diagram?
2. Why is there a difference in storage capacity between the logical disks and the physical disks?

Exercise 2: Planning a File Server and Operating System Configuration

The Lucerne Publishing network contains several additional file servers that must support a greater degree of fault tolerance than the basic file servers. For these specialized servers, the system partition must be protected and a small set of data files must be made highly available. You should also make the remaining user data fault tolerant using the most inexpensive method possible. To support this configuration, you plan to store the operating system on a 3-GB logical partition (partition C:), the highly available files on a 7-GB partition (partition D:), and the general user data on a 40-GB logical partition (partition E:). You plan to use RAID-1 and RAID-5 in combination with seven 10-GB hard disks.

On the diagram shown in Figure 3.6, you’ll need to label the logical partitions, the logical disks, and the physical disks. You should also include the size of each partition and disk, their functions, and the RAID configuration used for each logical disk.

Figure 3.6 - RAID configuration for a file server and operating system

1. How should you configure the data storage system for these servers?
2. Why is there a difference in storage capacity between the logical disk and the physical disks?

Exercise 3: Planning a Domain Controller and Services Configuration

The Lucerne Publishing network contains three Windows 2000 Server computers configured as domain controllers. The computers are also configured as DHCP and DNS servers. The servers must be highly available, although they don’t require a large amount of disk space. You plan to implement a RAID-1 configuration that uses two 10-GB hard disks and two 5-GB logical partitions.

1. Draw a diagram that provides a conceptual overview of how RAID will be implemented on these servers. Label the logical partitions, the logical disk, and the physical disks. Include the size of each partition and disk and their functions.

Exercise 4: Planning a Relational Database Server Configuration

Lucerne Publishing maintains several database servers. The RAID configuration should support the high availability needs of the operating system, the write- intensive needs of the database log files, and the large sequential reads of the database files. You plan to create three logical partitions: one for the operating system (partition C:), one for the database log files (partition D:), and one for the database files (partition E:). Partitions C: and D: will each be 10 GB. Partition E: will be 40 GB. Your configuration will use RAID-1 and RAID-5 to support three logical disks and nine 10-GB physical disks.

1. How should you configure the data storage system for these servers?
2. How much storage would the RAID-5 logical disk support if another 10-GB hard disk were added to the configuration?