IBM N Series, Руководство по оборудованию

Страница: 169 / 368

Chapter 10. Data protection with RAID Double Parity
149
10.2 Why use RAID-DP
Traditional single-parity RAID offers adequate protection against a single event. This event
can be a complete disk failure or a bit error during a read. In either event, data is re-created
by using parity data and data that remains on unaffected disks in the array or volume.
If the event is a read error, re-creating data happens almost instantaneously and the array or
volume remains in an online mode. However, if a disk fails, the lost data must be re-created.
The array or volume remains in a vulnerable degraded mode until data is reconstructed onto a
replacement disk or global hot spare disk. This degraded mode is where traditional
single-parity RAID fails to meet the demands of modern disk architectures. In single-parity
RAID, the chance of secondary disk failure is increased during rebuild times, which increases
the risk of data loss.
Modern disk architectures continue to evolve, as have other computer-related technologies.
Disk drives are orders of magnitude larger than they were when RAID was first introduced. As
disk drives become larger, their reliability did not improve and the bit error likelihood per drive
increased proportionally with larger media. These three factors (larger disks, unimproved
reliability, and increased bit errors with larger media) have serious consequences for the
ability of single-parity RAID to protect data.
Because disks are as likely to fail now as when RAID technology was first introduced, RAID is
still vital. Integrating RAID-DP when one disk fails, RAID re-creates data from both parities
and the remaining disks in the array or volume onto a hot spare disk. However, because RAID
was introduced, the significant increases in disk size resulted in much longer reconstruction
times for data that is lost on the failed disk.
It takes much longer to re-create lost data when a 274 GB disk fails than when a 36 GB disk
fails, as shown in Figure 10-2. In addition, reconstruction times are longer because the larger
disk drives that are used today tend to be ATA-based. ATA-based drives run more slowly and
are less reliable than smaller, SCSI-based drives.
Figure 10-2 Disk size versus reconstruction time