A system designed for optimal performance should have the filegroups/tablespaces spread across multiple sets of disks. The distribution of data across multiple channels is designed to balance disk I/O and maximize overall throughput when reading from and writing to the OnBase database.
As mentioned previously, it is very important to place the transaction log / redo log on its own set of disks (RAID 1+0). Additionally, it is recommended that the tempdb / temporary tablespace exists on its own set of disks. Every user request for a document hit list accesses both database files and the temporary space. This can cause contention when the two exist on the same disk drive, or when tempdb / temporary tablespace exists on the same drive as the transaction log / redo log.
It is recommended that you utilize smaller, faster disks (for example, 73 or 146 GB in size with 15K speed), particularly for the logs and temporary spaces.
The following layouts are recommended, and are based on the sets of disks that are available for the database installation.
These are basic recommendations that do not include custom requirements and sizing. Please contact your solution provider and/or the Database Services Group for more information. Recommendations for file layout can also be provided by the Database Services Group for existing implementations (where I/O patterns can be captured and analyzed).