Skip site navigation (1)Skip section navigation (2)
Date:      Sun, 22 Jan 2017 20:12:27 +0100
From:      Tobias Oberstein <tobias.oberstein@gmail.com>
To:        Daryl Richards <daryl@isletech.net>, freebsd-fs@freebsd.org
Subject:   Re: stuck on low scrub performance
Message-ID:  <4215534c-c249-beb2-e210-a5d31b0ac645@gmail.com>
In-Reply-To: <91c33947-bddc-014e-676d-2a5c3950bb71@isletech.net>
References:  <b0dc09d8-3579-f7c0-f72a-3e2ce798f443@gmail.com> <8b408926-768b-8df9-e38c-b617cce30b16@gmail.com> <91c33947-bddc-014e-676d-2a5c3950bb71@isletech.net>

next in thread | previous in thread | raw e-mail | index | archive | help
>> Note: the pool consists of 4 disk sets (striped) with 6 disks each
>> (RAID-Z2).
> Also, FWIW, I also find a scrub starts very slow for the first couple
> hours, then speeds up drastically. I also only have ~20TB so perhaps
> this scales?

I am wondering about this too, eg how large can one make a _single_ zpool?

Say I create a single zpool of N x 6 disks, where each 6er set is a RAID-Z2.

Currently, I have N=4 (with 6TB SAS disks) .. and seems works very well.

Will it work as well with N=16? Assuming I have enough SAS bandwidth of 
course ..

Oracle ZFS Storage ZS5-4 maxes out at 9PB.

Anyone knows about the zpool layout of this baby?

Cheers,
/Tobias




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4215534c-c249-beb2-e210-a5d31b0ac645>