UML软件工程组织

Raw Devices and Oracle - 20 Common Questions and Answers
本文摘自oracle官方网站

1. What is a raw device?

   A raw device, also known as a raw partition, is a disk partition that is

   not mounted and written to via the UNIX filesystem, but is accessed via

   a character-special device driver. It is up to the application how the

   data is written since there is no filesystem to do this on the

   application's behalf.

2. How can a raw device be recognised?

   In the '/dev' directory, there are essentially two type of files: block

   special and character special. Block special files are used when data is

   transferred to or from a device in fixed size amounts (blocks), whereas 

   character special files are used when data is transferred in varying 

   size amounts. Raw devices use character special files; a long listing

   of the '/dev' directory shows them with a 'c' at the leftmost position of

   the permissions field, e.g.

   crw-rw-rw-   1 root     system    15,  0 Mar 12 09:45 rfd0

   In addition, character special files usually have names beginning with 

   an 'r', as shown in the above example. Some devices, principally disks,

   have both a block special device and a character special device

   associated with them; for the floppy diskette shown above, there is also

   a device

   brw-rw-rw-   1 root     system    15,  0 Apr 16 15:42 /dev/fd0

   So the presence of a 'c' in a device does NOT necessarily mean this is a

   raw device suitable for use by Oracle (or another application).

   Generally, a raw device needs to be created and set aside for Oracle (or

   whatever application is going to use it) when the UNIX system is set

   up - therefore, this needs to be done with close cooperation between

   the DBA and UNIX system administrator.

   Once a raw device is in use by Oracle, it must be owned by the oracle

   account, and may be identified in this way.

3. What are the benefits of raw devices?

   There can be a performance benefit from using raw devices, since a write

   to a raw device bypasses the UNIX buffer cache, the data is transferred

   directly from the Oracle buffer cache to the disk. This is not guaranteed,

   though. If there is no I/O bottleneck, raw devices will not help. The

   performance benefit if there is a bottleneck can vary between a few

   percent to something like 40%. Note that the overall amount of I/O is

   not reduced; it is just done more efficiently.

   Another lesser benefit of raw devices is that no filesystem overhead

   is incurred in terms of inode allocation and maintenance or free block

   allocation and maintenance. 

4. How can I tell if I will benefit from using raw devices?

   There are two distinct parts to this: first, the Oracle database and

   application should be examined and tuned as necessary, using one or both

   of the following:

   -UTLBstat and UTLestat utilities (in $ORACLE_HOME/rdbms/admin)

   There are several strategies for improving performance with an existing

   disk arrangement, i.e. purely within Oracle. See [NOTE:16347.1] for

   details.

   After checking your Oracle database and application, the next stage is

   to identify UNIX-level I/O bottlenecks. This can be done using a UNIX

   utility such as 'sar' or 'vmstat'. See the relevant manual pages for

   details.

   If you identify that there is a UNIX-level problem with I/O, now is the

   time to start using raw devices. This may well require reorganisation of

   the entire UNIX  system (assuming there are no spare partitions

   available). 

5. Are there circumstances when raw devices have to be used?

   Yes. If you are using the Oracle Parallel Server, all data files,

   control files, and redo log files must be placed on raw partitions so

   they can be shared between nodes. This is a limitation with the UNIX

   operating system. Also, if you wish to use List I/O or Asynchronous I/O,

   some versions of UNIX require the data files and control files to be on

   raw devices for this to work. Consult your platform-specific

   documentation for details.

6. Can I use the entire raw partition for Oracle? 

   No. You should specify a tablespace slightly smaller in size than the

   raw partition size, specifically at least two Oracle block sizes

   smaller.

7. Can I use the first partition of a disk for a raw device?

   This is not recommended. On older versions of UNIX, the first partition

   contained such information as the disk partition table or logical volume

   control information, which if overwritten could render the disk useless.

   More recent UNIX versions do not have this problem as disk management

   is done in a more sophisticated manner. Consult your operating system

   vendor for more details, but if in any doubt do not use the first

   partition.

8. Who should own the raw device?

   You will need to create the raw devices as root, but the ownership

   should be changed to the 'oracle' account afterwards. The group must also

   be changed to the 'dba' group (usually called dba).

9. How do I specify a raw device in Oracle commands?

   When using a raw device you need to specify the full pathname in 

   single quotes, and use the REUSE parameter.  e.g. if there are two raw

   devices, each 30Mb in size, and the database has a 4K block size, the

   relevant command would look like this: 

   create tablespace raw_tabspace datafile '/dev/raw1' size 30712K REUSE

                                  datafile '/dev/raw2' size 30712K REUSE

10. Does the Oracle block size have any relevance on a raw device?

    It is of less importance than for a UNIX file; the size of the Oracle

    block can be changed, but it must be a multiple of the physical block

    size as it is only possible to seek to physical block boundaries and

    hence write only in multiples of the physical block size.

11. How can I back up my database files if they are on raw devices?

    You cannot use utilities such as 'tar' or 'cpio', which expect a filesystem

    to be present. You must use the 'dd' command, as follows:

    dd if=/dev/raw1 of=/dev/rmt0 bs=16k

    See the UNIX man page on dd for further details.

    It is also possible to copy the raw device file (using dd) to a normal 

    UNIX file, and then use a utility such as 'tar' or 'cpio', but this

    requires more disk space and has a greater administrative overhead.

12. Providing I am not using Parallel Server, can I use a mixture of raw 

    partitions and filesystem files for my tablespace locations?

    Yes. The drawback is that this makes your backup strategy more

    complicated.

13. Should I store my redo log files on raw partitions?

    Redo logs are particularly suitable candidates for being located on raw

    partitions, as they are write-intensive and in addition are written to

    sequentially. If Parallel Server is being used, redo logs must be

    stored on raw partitions.     

14. Can I use raw partitions for archive logs?

    No. Archive logs must be stored on a partition with a UNIX filesystem.

15. Can I have more than one data file on a raw partition?

    No. This means you should be careful when setting up the raw

    partition. Too small a size will necessitate reorganisation when you

    run out of space, whereas too large a size will waste any space the

    file does not use.

16. Should my raw partitions be on the same disk device?

    This is inadvisable, as there is likely to be contention. You should

    place raw devices on different disks, which should also be on different

    controllers. 

17. Do I need to make my raw partitions all the same size?

    This is not essential, but it provides flexibility in the event of

    having to change the database configuration.

18. Do I need to change any UNIX kernel parameters if I decide to use raw

    devices?

    No, but you may wish to reduce the size of the UNIX buffer cache if no

    other applications are using the machine.

19. What other UNIX-level changes could help to improve I/O performance?

    RAID and disk mirroring can be beneficial, depending on the application

    characteristics, especially whether it is read or write-intensive, or a

    mixture.

20. How can I gain further performance benefits, after considering all of the

    above?

    You will need to buy more disk drives and controllers for your system,

    to spread the I/O load between devices.


版权所有:UML软件工程组织