[OpenIndiana-discuss] ashift 13?

[Jim Klimov]

7 апреля 2015 г. 17:59:29 CEST, Jason Matthews <jason at broken.net> пишет:
>
>http://wiki.illumos.org/display/illumos/List+of+sd-config-list+entries+for+Advanced-Format+drives
>
>There are number of drives listed in the above document as being 8kb 
>sector devices. Is this right? ashfit=13?
>
>I set a system up on dc s3700 with a physical-block-size of 8192 and 
>most of the file systems i sent over reported approximately 2-3x more 
>disk usage than on the systems that were using the 512b advertised
>value.
>
>if 8k is properly aligned, does that mean the zpools built on 512byte 
>sectors, using the same drives, are under reporting their utilization?
>
>Can someone give me some color on what i am seeing? I am going to setup
>
>another system at 512b and send the seemingly over inflated file system
>
>back to see if it magically shrinks.
>
>thanks,
>j.
>
>
>
>_______________________________________________
>openindiana-discuss mailing list
>openindiana-discuss at openindiana.org
>http://openindiana.org/mailman/listinfo/openindiana-discuss

I think yes, for ssd's you can see larger block sizes like 8k. 

As for inflation - whenever you have smaller zfs allocations, such as those "tails" from blocks sized not power of 2 thanks to compression, they become a complete minimal "recordsize" block such as 4k or 8k as native to your drives, with trailing zero-padding. Some metadata blocks may also fall into this category, though for larger files these are clustered at 16kb(?) chunks. You also have less uberblocks in the fixed-size ring buffer of zpool labels.

On the upside, if your ssd does compression, these zeroes will in effect likely count toward wear-leveling reserves ;) With hdds this is more of a lost space as compared to 512-byte sectored disks. However this just become similar to usage on other systems (ext3, ntfs) typically with 4k clusters anyway. So we're told to not worry and love the a-bomb ;)

Jim
--
Typos courtesy of K-9 Mail on my Samsung Android