Based on my observation, most of OTLP ORA database’s block size is 4K,8K; VMFS’s block size is 1-8M, which depends on VMFS version and datastore capacity, the ratio of VMFS block size: Oracle Block size is over 250X. therefore I am concerning about IO efficiency and if Oracle IO in VM would generate lots of VMFS IO, which might cause some performance issues.
"When your Linux/Windows/Solaris guest reads/writes data, it does so in blocks and these blocks are usually around 64KB. For example, when your guest issues a SCSI read, then nothing happens to the VMFS at all - the read is "passed through" by the vmkernel straight to the SAN LUN.
VMFS does not get in the way of your guest read/writes, it does not do any kind of caching, the VMFS block size is irrelevant for guest I/O.
Where the VMFS block size is important is when fixing the maximum VMDK size"
dba_hba
63 Posts
868
0
Posted September 5th, 2011 02:00
Eddy
"The VMFS block size is not used for guest read/writes."
See this document
http://communities.vmware.com/docs/DOC-11458
under the section Understanding VMFS block size
and explained by the following paragraph
"When your Linux/Windows/Solaris guest reads/writes data, it does so in blocks and these blocks are usually around 64KB. For example, when your guest issues a SCSI read, then nothing happens to the VMFS at all - the read is "passed through" by the vmkernel straight to the SAN LUN.
VMFS does not get in the way of your guest read/writes, it does not do any kind of caching, the VMFS block size is irrelevant for guest I/O.
Where the VMFS block size is important is when fixing the maximum VMDK size"
Hope this helps
Allan