Return to MAIN-Index  Return to SUB-Index    IBM-AUSTRIA - PC-HW-Support    30 Aug 1999

RAID Functions



RAID Functions

Redundant Array of Independent Disks (RAID) technology gives you:


RAID technology stores data across groups of disk drives that are known as disk arrays. Subject to the level of RAID that you are using, this method of data storage provides the data redundancy that is needed for a secure system, and can allow faster retrieval of data through multiple channel access. Also, if a disk drive fails, you can normally exchange that disk drive without interruption of normal system operation.

The disk arrays can provide data redundancy that ensures that no data is lost if one disk drive in the array fails. The method that is used to write data to an array is related to the level of RAID that you are using. Disk arrays are contained in array subsystems. You can configure your subsystem with one or more arrays.

The SSA RAID Cluster Adapter provides RAID-1 facilities and also access to individual SSA disk drives that are not configured as members of an array.


RAID-1

RAID-1 is also known as mirroring or dual copy. It provides redundancy with better performance than a single disk, but requires real disk capacity of two times the data size.

A RAID-1 array has two disk drive members. If the members have different capacities, the resulting RAID-1 array has the capacity of the smaller member. Each data block of the array is written on each of the two members.

If one member fails, operations continue, to the good member. Performance is degraded for read operations (because read operations cannot now be shared between the two members), and improved for write operations (because only a single member is written). So a mixed workload may not show any change in overall performance. When a member returns or it is replaced (either manually or automatically by the hot-spare mechanism), a rebuild occurs. This rebuild is performed in parallel with any activity to the array.

Hot Spares

A hot-spare disk drive is a disk drive that is defined for automatic use if a disk drive within an array fails. The hot-spare must have a storage capacity greater than or equal to that of the smallest member of an array. You can define as many hot spares as you want. Any RAID-1 array on an adapter can use the hot-spare disk drives on that adapter. For full clustering operations to continue, the hot-spare used by a RAID-1 array must be in an SSA loop between the same two servers as the array.

If a disk drive within an array fails, after an interval, the adapter automatically uses a hot spare instead of the failed disk drive, and rebuilds the data that was on the failed disk on to the hot spare.


Back to  Jump to TOP-of-PAGE
More INFORMATION / HELP is available at the  IBM-HelpCenter

Please see the LEGAL  -  Trademark notice.
Feel free - send a Email-NOTE  for any BUG on this page found - Thank you.