Date: Thu, 04 Jul 2013 23:31:57 +0300 From: Volodymyr Kostyrko <c.kworr@gmail.com> To: Dmitry Morozovsky <marck@rinet.ru> Cc: freebsd-fs@freebsd.org, avg@freebsd.org Subject: Re: ZFS default compression algo for contemporary FreeBSD versions Message-ID: <51D5DBBD.70702@gmail.com> In-Reply-To: <alpine.BSF.2.00.1307042005120.2446@woozle.rinet.ru> References: <alpine.BSF.2.00.1307041620420.2446@woozle.rinet.ru> <51D576E1.6030803@gmail.com> <alpine.BSF.2.00.1307041950400.2446@woozle.rinet.ru> <51D59B6C.5030600@gmail.com> <alpine.BSF.2.00.1307042005120.2446@woozle.rinet.ru>
next in thread | previous in thread | raw e-mail | index | archive | help
04.07.2013 19:05, Dmitry Morozovsky wrote: > On Thu, 4 Jul 2013, Volodymyr Kostyrko wrote: > >>>>> is it sane to just set 'zfs compression=on dataset' to achieve best algo >>>>> on >>>>> fresh FreeBSD systems (-current and/or stable/9)? >>>> >>>> No and this is not safe AFAIK. Default compression is still lzjb and >>>> bootloader can't boot oof datasets compressed with lzjb. However on >>>> stable/9 >>>> you can simply set zfs compression=lz4 pool and everything would work fine >>>> if >>>> you updated the boot loader. >>> >>> I did not intend to compress root/boot datasets (and there is no much sense >>> in >>> this AFAICS); >>> >>> the second (and actually more important) my question is -- is lz4 in general >>> better than lzjb? >> >> Yes. Much better in terms of speed. > > Then, next logical step semms to me is to make lz4 the default ;-P As far as the code is too young and most other distribution are behind in terms of compatibility this is a no go. My naive dream is to see lz4hc in ZFS too. This way I can just give up at compressing logs. -- Sphinx of black quartz judge my vow.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?51D5DBBD.70702>