CIS 170F: Windows 7 Administration

Week 4

Managing Disks
Disk Management Tasks
Managing Fault Tolerance

  • Basic disks are not fault tolerant by design. If the data is not backed up, the loss of a basic disk will result in permanent data loss.
  • Windows 2000 Server, Windows Server 2003, and Windows Server 2008 are server operating systems that support fault-tolerant dynamic volumes. The following two types of fault-tolerant volumes are supported by dynamic disks:
    • Mirrored volumes: They consist of identical data mirrored across two dynamic disks. Versions of Windows 7 that support dynamic disks only support mirrored volumes.
      • If a single disk fails in a mirrored set, the mirror can be broken using the Disk Management console or the DiskPart command-line utility. Breaking a mirror means that the remaining disk is told that it no longer has a partner to replicate with. The volume that was being mirrored reverts to just a simple volume. Once a replacement disk is added back into the computer as a dynamic disk, the simple volume can be converted back into a mirrored volume.
      • If both disks fail in a mirrored set, such as when a common component like a data cable fails, then the mirrored volume is unavailable. If neither disk can be revived, the mirror set has to be rebuilt from scratch and the volume's data is then restored from a backup copy.
    • RAID 5 volumes: They consist of striped data and parity information across three to 32 dynamic disks.
      • If a single disk fails in a RAID-5 array of disks, the RAID-5 volume will continue to operate in a degraded mode. A replacement dynamic disk drive must be added to the computer and a repair operation initiated with the Disk Management console or the DiskPart command-line utility.
      • If more than one disk fails in a RAID-5 array of disks, the entire RAID-5 volume becomes unavailable. In this case, the RAID-5 array would have to be rebuilt from scratch and the data restored from a backup copy.
      • Hardware RAID is much faster and easier when a hard disk fails. Many hardware RAID systems are hot-swappable, meaning that when a disk fails, you just slide it out, insert a new disk, and RAID gets back to work. This is not the case with software RAID. Recovering from a single hard-disk failure can take time, which in turn costs an organization money. The downside to hardware RAID is cost. A server that has hardware RAID built in can cost thousands of dollars. Software RAID is free.