From mboxrd@z Thu Jan 1 00:00:00 1970 From: Gregory Leblanc Subject: Re: Soft-/Hardware RAID Performance Date: 20 Feb 2003 10:31:38 -0800 Sender: linux-raid-owner@vger.kernel.org Message-ID: <1045765895.2116.2.camel@gregdell> References: <5.1.0.14.0.20030220112347.028ef5c0@195.90.31.33> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=pgp-sha1; protocol="application/pgp-signature"; boundary="=-JHpiVSQhP7ncU3t7xgcp" Return-path: In-Reply-To: <5.1.0.14.0.20030220112347.028ef5c0@195.90.31.33> To: linux-raid@vger.kernel.org List-Id: linux-raid.ids --=-JHpiVSQhP7ncU3t7xgcp Content-Type: text/plain Content-Transfer-Encoding: quoted-printable On Thu, 2003-02-20 at 02:55, Daniel Brockhaus wrote: [snip] > One suggestion to speed up the reads was to issue several reads in=20 > parallel. Silly me didn't think of that, I was completely focused on the=20 > writes, which are more important for my application. Anyway. Using parall= el=20 > reads (from four processes), read performance scales almost with the numb= er=20 > of disks in the array. This goes for both, hardware and software RAID, wi= th=20 > software RAID being about 15% faster. >=20 > Write performance on the other hand does not change at all when using=20 > multiple processes - for obvious reasons: The kernel queues, sorts and=20 > merges write requests anyway, so the number of processes doing the writes= =20 > does not matter. But I've noticed something peculiar: If I change my=20 > benchmark to write 4K blocks at 4K boundaries, write performance increase= s=20 > to almost 300%. This is quite logical, since the kernel can write a 'page= =20 What was your previous benchmark write size? And what parameters did you use when creating the RAID 0 array and filesystem? There may be opportunities for tuning there. Greg [snip] --=-JHpiVSQhP7ncU3t7xgcp Content-Type: application/pgp-signature; name=signature.asc Content-Description: This is a digitally signed message part -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux) iD8DBQA+VR8GnAkVXO6fw4sRAnKqAJ9ksAQ2Lxf8+GEiuhruPmX9PWg8WwCeL3x1 uJFklugLBG1SXYyGmeNAhgI= =c2gA -----END PGP SIGNATURE----- --=-JHpiVSQhP7ncU3t7xgcp--