[linux-audio-user] Submitted for your approval

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On Mon, Apr 07, 2003 at 09:13:26AM -0700, R Parker wrote:
> I recently asked about this on the jack and ardour
> developers list. I think Jan replied by stating that
> he thought jackd might be reading the period buffer
> and compensating accordingly--if "-p 512" then move
> file to compensate for the inherint latency of that
> period.

jackd itsself doesnt do it, but ardour can correct automatically given the
information that jackd provides, I think.
 
> I've got to setup a session but will come back to
> learn more about the RAID latency issues that you're
> outlining. BTW, I'm running RAID 5 but have thought 10
> would be optimal for performance but expensive.

Coincidentally I was talking with a sysprog over coffee about raid 10 and
he said that the survivability of raid 10 was not as good as 5, there are
relativly likely disk loss patterns that can cause you to loose all the
data, when a raid 5 setup with the same (or less) redundancy could
survive. He mentioned it cos it actually happened over the weekend :(

I dont know how the performance compares. I've always used 0 for raw
throughput and terrible survivability, or 5 for robustness and a bit extra
speed.

- Steve


[Index of Archives]     [Linux Sound]     [ALSA Users]     [Pulse Audio]     [ALSA Devel]     [Sox Users]     [Linux Media]     [Kernel]     [Photo Sharing]     [Gimp]     [Yosemite News]     [Linux Media]

  Powered by Linux