HSG80 ACS Solution Software V8.6 for IBM AIM Installation and Configuration Guide

Planning Storage 2–19
RAIDsets are similar to stripesets in that the I/O requests are broken into smaller “chunks”
and striped across the disk drives. RAIDsets also create chunks of parity data and stripe
them across all the members of the RAIDset. Parity data is derived mathematically from
the I/O data and enables the controller to reconstruct the I/O data if a single disk drive
fails. Thus, it becomes possible to lose a disk drive without losing access to the data it
contained. Data could be lost if a second disk drive fails before the controller replaces the
first failed disk drive and reconstructs the data.
The relationship between the chunk size and the average request size determines if striping
maximizes the request rate or the data-transfer rates. You can set the chunk size or use the
default setting. See “Chunk Size,” page 2–26, for information about setting the chunk size.
Keep these points in mind when planning RAIDsets:
■ Reporting methods and size limitations prevent certain operating systems from
working with large RAIDsets.
■ Both cache modules must be the same size.
■ A RAIDset must include at least 3 disk drives, but no more than 14.
■ A storageset should only contain disk drives of the same capacity. The controller limits
the capacity of each member to the capacity of the smallest member in the storageset.
Thus, if you combine 9 GB disk drives with 4 GB disk drives in the same storageset,
you waste 5 GB of capacity on each 9 GB member.
■ RAIDsets are particularly well-suited for the following:
❏ Small to medium I/O requests
❏ Applications requiring high availability
❏ High read request rates
❏ Inquiry-type transaction processing
■ RAIDsets are not particularly well-suited for the following:
❏ Write-intensive applications
❏ Database applications in which fields are continually updated
❏ Transaction processing