Users Guide

virtual disk, the controller must decide which free space on the physical disks to allocate to the new virtual disk. The PERC controllers

look for the largest area of free space and allocate this space to the new virtual disk.

• SCSI limitation of 2TB — Virtual disks created on a PERC controller cannot be created from physical disks with an aggregate size

greater than 2TB. This is a limitation of the controller implementation. For example, you cannot select more than 30 physical disks that

are 73GB in size, regardless of the size of the resulting virtual disk. When attempting to select more than 30 disks of this size, a pop-up

message is displayed indicating that the 2TB limit has been reached, and that you should select a smaller number of physical disks. The

2TB limit is an industry-wide SCSI limitation.

• Expanding virtual disks — You can only use the Recongure task to expand a virtual disk that uses the full capacity of its member

physical disks.

• Reconguring virtual disks — The Recongure task is not available when you have more than one virtual disk using the same set of

physical disks. You can, however, recongure a virtual disk that is the only virtual disk residing on a set of physical disks.

• Virtual disk names not stored on controller — The names of the virtual disks that you create are not stored on the controller. If you

reboot using a dierent operating system, the new operating system may rename the virtual disk using its own naming conventions.

• Creating and deleting virtual disks on cluster-enabled controllers — There are particular considerations for creating or deleting a virtual

disk from a cluster-enabled controller.

• Implementing channel redundancy — A virtual disk is channel-redundant when it maintains redundant data on more than one channel.

If one of the channels fails, data is not lost because redundant data resides on another channel.

NOTE: For more information about channel redundancy, see Channel Redundancy And Thermal Shutdown.

• Rebuilding data — An failed physical disk that is used by both redundant and nonredundant virtual disks cannot be rebuilt. Rebuilding a

failed physical disk in this situation requires deleting the nonredundant virtual disk.

• Disk group concept consideration for S110 — Disk grouping is a logical grouping of disks attached to a RAID controller on which one or

more virtual disks are created, such that all virtual disks in the disk group use all of the physical disks in the disk group. The current

implementation supports the blocking of mixed disk groups during the creation of logical devices.

Physical disks are bound to disk groups, therefore, there is no RAID level mixing on one disk group.

Storage Management Server implements the disk group concept during virtual disk creation. Functionally, after a group of physical disks is

used to create their rst virtual disk, unused space in the disk is used only to expand the virtual disk, or create new virtual disks in the

unused space. The virtual disks have identical RAID level.

Also, existing mixed conguration is not aected. However, you cannot create mixed congurations.

You can read or write to the virtual disks, rebuild, and delete the disks.

You cannot create virtual disks on a set of disks migrated from earlier software RAID versions and congured with multiple RAID levels.