[Date Prev][Date Next]
Re: [ldap] Implementation Suggestions
Igor Brezac wrote:
I wouldn't expect the noatime option to make much difference; though
using direct I/O in BerkeleyDB could help a little. It's important to
remember that the BDB cache files aren't read or written like normal
files, they're just serving as backing store for a mapped memory region.
BDB won't actually issue reads or writes to the individual database
files unless it needs to flush the BDB cache or pull something in that's
not in the cache. Solaris will generally only update the atime once
after the first mmap reference.
On Fri, 20 Jan 2006, Spicer, Kevin wrote:
From: Daniel Henninger [mailto:email@example.com]
It may be worth adding the noatime flag to the filesystems that hold
data and (bdb) logs. You are indexing a lot of attributes, do you
No atime? That's interesting. I've never heard of that, what does
By default, the file system is mounted with
normal access time (atime) recording. If
noatime is specified, the file system will
ignore access time updates on files, except
when they coincide with updates to the
ctime or mtime. See stat(2). This option
reduces disk activity on file systems where
access times are unimportant (for example,
a Usenet news spool).
noatime turns off access time recording
regardless of dfratime or nodfratime.
The POSIX standard requires that access
times be marked on files. -noatime ignores
them unless the file is also modified.
I think this was introduced in Sol7 or 8,
This is still not a lot files to really make a difference.
You are better off looking at forcedirectio if it is available on
solaris 8. This certainly helps with sql db performance.
-- Howard Chu
Chief Architect, Symas Corp. http://www.symas.com
Director, Highland Sun http://highlandsun.com/hyc
OpenLDAP Core Team http://www.openldap.org/project/