Re: ASM is single point of failure ?

  • From: Tanel Põder <tanel.poder.003@xxxxxxx>
  • To: <kevinc@xxxxxxxxxxxxx>, "ORACLE-L" <oracle-l@xxxxxxxxxxxxx>
  • Date: Tue, 9 May 2006 21:35:50 +0800

Hi,

ASM is not in the I/O code path. Huh, what did he say? Oracle
positions and performs reads and writes and that is done with
either libC, LibODM or if you are a real experimental type ASMLib (on
Linux). There are no comms between a shadow process and an ASM process
for I/O (e.g., db file sequential read, direct path write, etc). There
is for metadata ops, such as adding a datafile, or adding an extent,
dropping a datafile, etc...

Btw, one case when ASM can be in IO codepath is when the ASM allocation unit -> physical location in datafile isn't cached in SGA for a particular extent or block range. When trying to access a block in this range, Oracle server process makes an OCI call to ASM instance to find out where a range of blocks is physically located. ASM instance gives the answer and Oracle DB instance caches this info in SGA for future use.


As long as the info is cached in local SGA and no metadata changes are required, ASM instance is out ot of the way of IO codepath.

Tanel.


-- //www.freelists.org/webpage/oracle-l


Other related posts: