Date: Sat, 12 May 2007 13:04:28 -0700 From: Bakul Shah <bakul@bitblocks.com> To: Peter Schuller <peter.schuller@infidyne.com> Cc: freebsd-current@freebsd.org, efinley@efinley.com Subject: Re: ZFS the perfect FS? if only... Message-ID: <20070512200428.E996A5B3E@mail.bitblocks.com> In-Reply-To: Your message of "Sat, 12 May 2007 09:29:12 %2B0200." <46456CC8.30909@infidyne.com>
next in thread | previous in thread | raw e-mail | index | archive | help
> > zpool create foo raidz2 dev0 dev1 dev2 dev3 > > zpool add foo raidz2 dev4 dev5 dev6 dev7 > > > > The new devices are in a different raidz2 group but but *all > > of the space* will be used for any filesystem on this pool -- > > isn't this good enough and if not, why? > > Isn't it obvious? > > You waste less diskspace. In the above setup you are only using > half your diskspace and depending on what you need it for you might > aswell have been using a mirror... > > Often having two parity drives can be more important than the ratio of > parity vs. non-parity. Fair enough. > Also, if you DO want 50% parity, a single raidz2 with 8 drives with 50% > parity is more resilient to failures than 2x4 raidz2 since any 4 drives > can fail whereas in the latter case any 2 drives can fail, or up 4 > drives if they happen to be the right drives. I believe raidz2 means two parity blocks so if you want 50% parity you'd need raidz4. But that nit aside, you have a valid point. > > Not worth it > > Most definitely worth it in many situations where performance is just > not the goal. So what if it takes a week to perform the operation, as > long as the array is not degraded during this timewindow. It is not obvious at all that performance would not be a goal for a freebsd user! It *is* obvious that you would want more space but not obvious how to do the conversion from an N disk raidz2 array to N+1 disk raidz2 array *without bringing the whole array down*. Even you may care about the array being down for hours/days! Thinking more about this, I believe this can be done without adding too many complications. Proof left as an exercise:-)
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20070512200428.E996A5B3E>