A Windows 2000 server has five disks; four 100-GB disks currently form a single striped (RAID 0) volume. You want to reconfigure so the volume is fault-tolerant while preserving as much usable space as possible, using only existing hardware. What is the correct approach?

Introduction / Context:
RAID-0 (striping) offers performance but no redundancy: a single disk failure destroys the volume. To add fault tolerance while maximizing capacity across four equal-sized disks, RAID-5 (striping with distributed parity) is the typical choice on Windows 2000 dynamic disks, sacrificing the capacity of one disk for parity.

Given Data / Assumptions:

• Four 100-GB disks currently in a non-fault-tolerant stripe.
• Goal: introduce fault tolerance using existing disks.
• Also maximize usable space.

Concept / Approach:

Windows cannot convert a live RAID-0 directly to RAID-5. The supported path is: back up data, delete the RAID-0, create a new RAID-5 across the four disks, restore data. RAID-5 usable capacity = (N − 1) * size = (4 − 1) * 100 GB = 300 GB, providing single-disk fault tolerance. A mirror (RAID-1) would use pairs and yield only 200 GB across four disks (two mirrors), which is less capacity.

Step-by-Step Solution:

Verification / Alternative check:

After creation, Disk Management will report a healthy RAID-5 volume with 300 GB usable and parity distributed. Simulate a disk failure (pull one disk in a lab) to verify continued access and rebuild capability.

Why Other Options Are Wrong:

Option A: Incomplete and does not create fault tolerance.

Option C: Mirroring reduces capacity to 200 GB; not “as much as possible.”

Option D: Stops short of specifying the new, fault-tolerant configuration.

None: A clear, supported approach exists (RAID-5).

Common Pitfalls:

Expecting in-place conversion; underestimating rebuild time and ensuring verified backups before destructive changes.

Final Answer:

Back up the data, delete the stripe, create a RAID-5 volume across the four disks, then restore the data