Software User Guide for Windows*

RAID Features
Intel
®
RSTe for Microsoft Windows* OS Software User’s Guide 11
2.5.20 Dirty Stripe Journaling
Intel
®
RSTe provides support for Dirty Stripe Journaling (DSJ). DSJ is used to help speed up RAID 5 write
power loss recovery by storing the write stripes that were in progress at the time of the failure. The DSJ
allows rapid recovery without having to rebuild the entire volume. The DSJ is only utilized when disk write
cache is DISABLED.
2.5.21 Partial Parity Logging (PPL)
Intel
®
RSTe provides support for Partial Parity Logging (PPL). PPL is used to record the results of XORing
old data with old parity. PPL is currently saved as part of the RAID member information and is only
utilized when writing RAID 5 parity. It helps protect against data loss when a power failure or a system
crash occurs by allowing data to be rebuilt by utilizing the PPL information.
2.5.22 OS Installation
Intel
®
RSTe provides the OS appropriate RSTe driver files required for installation during the OS setup
onto a drive or RAID volume attached to either the AHCI or SCU controllers.
2.5.23 Selectable Boot Volume
Intel
®
RST 3.0 supports the ability to select any volume as the OS boot volume. The OS installer can install
the operating system onto RAID volume. There is no need for RAID management (for example, volume
creation/deletion) support from within OS installer.
2.5.24 Auto Rebuild
Intel
®
RSTe provides support for the ability to automatically rebuild a failed or degraded RAID volume.
This feature will begin when a member disk of the array has failed and a suitable replacement disk with
sufficient capacity is available. As soon as the failure occurs the rebuild process will begin automatically,
using the marked Hot Spare disk, without user intervention.
If a marked Hot Spare disk is not present, the automatic rebuild process will begin under the following
conditions:
 Another free disk is plugged into the same directly attached physical location as the failed drive.
 The newly inserted disk size is at least as large as the amount of space used per disk in the current
array.
 The newly inserted disk must be the same type (SAS/SATA) as the disk being replaced or the rebuild
will not start..
 If the newly inserted disk contains Intel
®
RSTe (or Intel
®
RST) metadata with current status of member
being offline or contains no Intel
®
RSTe (or Intel
®
RST) metadata.
 The newly inserted disk has not reported a SMART event.
The following table summarizes the functionality.
Table 2. Auto Rebuild
Controller
Auto Rebuild Support
Action
AHCI and SCU
Previously marked Hot Spare
available.
Rebuild starts when spare
found. This takes precedence
over auto-spare disk.
AHCI
No Hot Spare previously
marked.
No auto rebuild: Manual
steps required to rebuild
array using new disk.
SCU
Auto rebuild conditions
described above are met.
Auto rebuild starts without
any user intervention.