|
 |
 |
To access the contents, click the chapter and section titles.
Oracle Performance Tuning and Optimization
(Publisher: Macmillan Computer Publishing)
Author(s): Edward Whalen
ISBN: 067230886x
Publication Date: 04/01/96
Partial Data Protection
Another option for protecting your data is to apply fault tolerance to specific pieces of critical data. This may be the best option if you have a limited budget and require high performance.
Critical areas that should be protected are the OS, the redo logs, and the archive logs. If these areas are protected and a drive volume failed, you would not have to spend extra time restoring or reloading the operating system. If the redo log volumes are not protected and a drive failed, you might not be able to recover your database to the point of failure. If the archive files are not protected and a drive failed, you might not be able to recover in the event of drive failure on a data volume.
By using partial fault tolerance, you may be able to reduce downtime in the event of a failure and still keep within your budget. By protecting areas such as the OS volumes and redo log volumes, you can reduce the amount of time needed to recover.
Fault Tolerance Summary
The most likely component in your system to fail is a disk drive because it is a heavily used mechanical device. To avoid extended downtime, it is wise to employ some type of fault tolerance. The extent to which you protect your data depends on your uptime requirements, performance requirements, and budget.
Depending on your needs, you may not have to completely protect all your databut some fault tolerance is recommended. If it is too expensive to protect all your disks, protect at least the critical areas. Depending on your needs, you may want to use different volumes. Here are a few suggestions about fault tolerance for you to consider:
- Fully protect your OS and Oracle binaries using RAID-1 or RAID-5. This level of protection is worth the cost of the extra disks because you avoid reloading the OS and Oracle.
- Fully protect the redo log files. If possible, put each redo log file on a separate mirrored volume. By separating the redo log files, the sequential nature of the I/Os generated by the LGWR is maintained even when archiving.
- Use RAID-5 on read-intensive volumes. The performance degradation experienced when using RAID-5 occurs only on writes. If the volume is mainly used for reads, very good performance can be achieved with RAID-5.
The amount and type of fault tolerance you employ depends on your specific requirements. If you cannot permit any downtime, you must protect yourself against disk failures.
|
Configuring RAID for RDBMS Performance
RAID volumes can be very useful in an RDBMS system. One key advantage of RAID is its ability to spread I/Os evenly across a large number of disks. If a disk array has a 16K stripe, roughly every 8 database blocks will be on a different disk.
If you do random I/Os, chances are good that the I/Os will be fairly evenly spread across the disks. By creating a disk array with a sufficient number of disk drives, I/Os should not be a problem.
Here are some configuration issues you should consider:
- Isolate sequential I/Os. Even in a RAID configuration, drives are driven sequentially if the I/Os to the logical volume are sequential. Isolate areas such as the redo log files and log archive destination. By separating the sequential I/Os, you can drive the disks used for the sequential volumes at a much higher I/O rate.
- Distribute random I/Os. A larger number of disks in a random volume can support a larger number of random I/Os. The more disks you have in the volume, the more random I/Os you can support.
- Size the volumes properly. An underconfigured volume causes an I/O bottleneck. Dont exceed I/O limitations. In Chapter 14, Advanced Disk I/O Concepts, you learned how to determine the optimum number of I/Os per second per disk. Be sure to take into account the extra I/Os generated by the disk array for fault tolerance.
- Configure for the disk array. To take maximum advantage of disk array technology, you must abandon some of the configuration guidelines you have used in the past. Some of these new ideas are the complete opposite of what you have done in the past.
Because all these issues are important, the following sections look at each individually.
|