[BBLISA] Moving from RAID 0 to LVM RAID?

[Daniel Hagerty]

"Edward Ned Harvey" <bblisa2 at nedharvey.com> writes:

> #2  Not based on cpu strain, better performance.  I believe every OS will
> write files, and in the software level, the kernel will not tell the
> application that a write has finished, until the write has finished.  At

    It's more complicated than that.  The kernel is allowed to
claim write completion without having pushed the data all the way
through to the platter w/ an acknowledgement.  You would be very
unhappy if it defaulted the other way.

    close() is guaranteed to have flushed all outstanding writes
before it returns.  fsync does the same without actually closing the
file.  sync generally does this for all outstanding writes on the
system, but in practice it's the wrong tool for almost every job.