From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mari.romanrm.net ([157.7.203.202]:37955 "EHLO mari.romanrm.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753075AbaGTTgR (ORCPT ); Sun, 20 Jul 2014 15:36:17 -0400 Date: Mon, 21 Jul 2014 01:36:09 +0600 From: Roman Mamedov To: Bob Marley Cc: TM , linux-btrfs@vger.kernel.org Subject: Re: 1 week to rebuid 4x 3TB raid10 is a long time! Message-ID: <20140721013609.6d99c399@natsu> In-Reply-To: <53CC1553.1020908@shiftmail.org> References: <53CC1553.1020908@shiftmail.org> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=PGP-SHA1; boundary="Sig_/4/vEyKgwSQ8RubYmSwW8tbi"; protocol="application/pgp-signature" Sender: linux-btrfs-owner@vger.kernel.org List-ID: --Sig_/4/vEyKgwSQ8RubYmSwW8tbi Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Sun, 20 Jul 2014 21:15:31 +0200 Bob Marley wrote: > Hi TM, are you doing other significant filesystem activity during this=20 > rebuild, especially random accesses? > This can reduce performances a lot on HDDs. > E.g. if you were doing strenous multithreaded random writes in the=20 > meanwhile, I could expect even less than 5MB/sec overall... I believe the problem here might be that a Btrfs rebuild *is* a strenuous random read (+ random-ish write) just by itself. Mdadm-based RAID would rebuild the array reading/writing disks in a complet= ely linear manner, and it would finish an order of magnitude faster. --=20 With respect, Roman --Sig_/4/vEyKgwSQ8RubYmSwW8tbi Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.12 (GNU/Linux) iEYEARECAAYFAlPMGikACgkQTLKSvz+PZwhHxwCcDTfZt26R1n6D1gsZHnqN/P+8 fYgAn0aMKMgZBVRRB16nrqfzceXz/Pb7 =u7m8 -----END PGP SIGNATURE----- --Sig_/4/vEyKgwSQ8RubYmSwW8tbi--