4.1 Background InformationRedundant Arrays of Independent Disks (RAID) is a storage technology used to improve the processing capability of storage systems. This technology is designed to provide reliability in disk array systems and to take advantage of the performance gains multiple disks can offer.
RAID comes with a redundancy feature that ensures fault-tolerant, uninterrupted disk storage operations. In the event of a disk failure, disk access will still continue normally with the failure transparent to the host system.
RAID has six levels: RAID 0 ~ 5. RAID levels 1, 3 and 5 are the most commonly used levels, while RAID levels 2 and 4 are less popular. Appendix D, RAID Levels, gives information about these levels, including the benefits of each.
Infortrend disk array controllers support hot-swapping where a failed drive can be replaced while the disk array system continues to function. Spares can also be assigned so that, as soon as a drive fails, the spare will be automatically configured into the array and reconstruction will commence.
4.1.1 Definition of TermsThis section describes some of the disk array terms used in this documentation.
Physical drives. These are the actual SCSI drives installed on the connectors of the SCSI cables. These drives are displayed in Physical View under the RAID View window.
- Spare drives. These are physical drives that serve as backups. When a drive fails, the spare is automatically configured into the array, and data reconstruction will commence immediately. Spare drives appear in darker (shaded) colors than normal drives and have a red cross superimposed on them. Large red crosses indicate Global spares, smaller ones represent Local spares.
- Replacement drives. These are physical drives that are manually configured into the array to replace failed drives. In the absence of spare drives, you will need to use replacement drives to replace defective drives before rebuilding. If a spare drive has been used to rebuild the array, you will also need to replace the failed drive manually to create another spare with the precaution that another drive might fail.
- Failed drives. These are physical drives that fail due to some type of error. Failed drives appear with large red X marks on their respective icons.
- Logical drives. These drives are created using physical drives. Combining physical drives into logical drives gives you a disk array with a certain RAID level. To view logical drives, use Logical View under the RAID View window.
- Logical volumes. These volumes are created using logical drives. Combining logical drives into logical volumes gives you a single logical unit with even larger capacity. Logical volumes or their partitions are mapped to various host LUNs. To view logical volumes, use Logical View under the RAID View window.
4.1.2 Operating With Spare DrivesYou can assign spare drives to a particular logical drive to serve as backup drives. When a drive fails within the logical drive, one of the spares will be automatically configured into the logical drive, and data reconstruction onto it will immediately commence.
The following are guidelines for disk failure recovery when a spare drive is available:
- If a spare drive exists in the same logical drive, the controller will automatically mount the spare drive and start data rebuilding in the background.
- Depending on the design of the system external to the controller, it may be possible to remove a defective drive and replace it with a new drive without shutting down the system (hot-swapping). Alternatively, the system can be shut down at a convenient time and the failed drive replaced.
- The replacement drive must then be assigned as a new spare drive.