Summary

This chapter started by covering the essentials of RAID volumes. RAID 0 volumes are commonly implemented as striped volumes and are used to increase disk I/O. Solaris also supports RAID 0 concatenated volumes. RAID 1 volumes are mirrored volumes. They provide data redundancy, but at the cost of using double the disk space. RAID 5 volumes provide both striping and data redundancy.

The Solaris Volume Manager enables you to create and manage storage volumes in Solaris 9. Through the command line, SVM is accessed through a series of meta* commands, such as metainit, metastat, and metadb. The Solaris Management Console also provides an interface for Solaris Volume Manager.

Through SVM, you can create RAID 0, RAID 1, and RAID 5 volumes. You can also create soft partitions, which enable you to exceed the rigid limit of eight slices per hard disk. You can also create transactional volumes, although these are being phased out in favor of UFS logging.

SVM also enables you to create hot spare pools, which are a series of backup slices provided in case of a RAID 1 or RAID 5 component failure. For computers that require high availability, hot spares are essential. Disk sets can also be created with SVM, but to get the full functionality out of using disk sets (such as failover), you will need an additional high availability management product.