file system journaling (was: Re: i-ram vs. tmpfs (was: Re: Mail clustering))
On [Fri, 13.04.2007 12:06], Anders Breindahl wrote:
AB> On 200704101607, Michael Loftis wrote:
AB> > RAID-5 is extremely slow at writes. Also can't forget about seeks.
AB> > Depending on your access patterns an external journal can be a definite
AB> > win, sometimes it's always a win with certain access patterns. EXT3 also
AB> > by default only journals metadata, not data, which means the write rate to
AB> > the journal is relatively low.
AB> Allow me to correct:
AB> Ext3 journals both data and metadata. Xfs only journals metadata. But
AB> that might be a relevant issue here, too, as many seem to prefer xfs on
AB> their mail servers.
no, that's not correct. ext3 can do both, jounral only metadata or
metadata+data, although i don't know what's the default.
see "man tune2fs":
When the filesystem is mounted with journalling enabled, all data (not just metadata) is committed
into the journal prior to being written into the main filesystem.
When the filesystem is mounted with journalling enabled, all data is forced directly out to the main
file system prior to its metadata being committed to the journal.
When the filesystem is mounted with journalling enabled, data may be written into the main filesys-
tem after its metadata has been committed to the journal. This may increase throughput, however, it
may allow old data to appear in files after a crash and journal recovery.