Return-Path: linux-nfs-owner@vger.kernel.org Received: from cantor2.suse.de ([195.135.220.15]:50638 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751424Ab3GOEfi (ORCPT ); Mon, 15 Jul 2013 00:35:38 -0400 Date: Mon, 15 Jul 2013 14:35:27 +1000 From: NeilBrown To: Ben Greear Cc: Dean , "J.Bruce Fields" , Olga Kornievskaia , NFS Subject: Re: Is tcp autotuning really what NFS wants? Message-ID: <20130715143527.4eca7283@notabene.brown> In-Reply-To: <51DD9C5A.3000505@candelatech.com> References: <20130710092255.0240a36d@notabene.brown> <20130710022735.GI8281@fieldses.org> <51DD9AD5.1030508@gmail.com> <51DD9C5A.3000505@candelatech.com> Mime-Version: 1.0 Content-Type: multipart/signed; micalg=PGP-SHA1; boundary="Sig_/g3.94vV3oovgjsBx/2YeQpu"; protocol="application/pgp-signature" Sender: linux-nfs-owner@vger.kernel.org List-ID: --Sig_/g3.94vV3oovgjsBx/2YeQpu Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: quoted-printable On Wed, 10 Jul 2013 10:39:38 -0700 Ben Greear wro= te: > On 07/10/2013 10:33 AM, Dean wrote: > > > This could significantly limit the amount of parallelism that can be= achieved for a single TCP connection (and given that the > > > Linux client strongly prefers a single connection now, this could be= come more of an issue). > > > > I understand the simplicity in using a single tcp connection, but perfo= rmance-wise it is definitely not the way to go on WAN links. When even a mi= niscule amount > > of packet loss is added to the link (<0.001% packet loss), the tcp buff= er collapses and performance drops significantly (especially on 10GigE WAN = links). I > > think new TCP algorithms could help the problem somewhat, but nothing a= vailable today makes much of a difference vs. cubic. > > > > Using multiple tcp connections allows better saturation of the link, si= nce when packet loss occurs on a stream, the other streams can fill the voi= d. Today, the > > only solution is to scale up the number of physical clients, which has = high coordination overhead, or use a wan accelerator such as Bitspeed or Ri= verbed (which > > comes with its own issues such as extra hardware, cost, etc). >=20 > I have a set of patches that allows one to do multiple unique mounts to t= he same server from a single > client, but the patches are for the client side, so it would not help > non-Linux clients. And, the patches were rejected for upstream as not be= ing > useful. But, if you are interested in such, please let me know and I can= point > you to them... Yes please! NeilBrown --Sig_/g3.94vV3oovgjsBx/2YeQpu Content-Type: application/pgp-signature; name=signature.asc Content-Disposition: attachment; filename=signature.asc -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.19 (GNU/Linux) iQIVAwUBUeN8Dznsnt1WYoG5AQIJnw/+NX2XltL5HwgcZncP9eTGIpPs1hXMxl0L HjHwlTk5dvN4J8YtVYLVKYswY5DtAtakbM+oYODLTXrnwjB4zS2rghm9M0azgzAH c8b9o8XoGAQnGnPos+357HqcN7v8/EOyyrFLAzj1h6++XXn9QSpGzhVVq8LBURlu qc5tjGe5DEf5SnELPZFQ8ZI6hdN0RRfr0X3YzwCzbLiUaC+bIwjYzye3tFLBl5hQ Io/PWLraR4Jw7bWsuGdPxVyHPs99gHkbUKUmP1tEUVO6pJU3XbWdu+hXJUEFPqYX ZNRhwkgu2kynuOi1U5XTUlXo7V1SPI9Z5zcOjvmkYY+M1QQNN+t0U/vrDauhu73r MSu2w/pvHb12ZWV3dhuJxAsJNsWwXh2El1dSpLJGLW1UE287DqPhlTidB+LCdr1y bZmiNQNCnabdk2fPSJgfEwC9FcrdL7+OlXyHkrQTLoS/61Kr+tNvUfmaY8egIpai YMLnBalMJMIvPQc46QykNAzZPs5w4gy22pT62+xnQ51Hm1W5anKsv3qez7OvNxg4 CWOcSCMU/dnjm47GUSA7x5UPY24+uVF/gtCrxRSbZxLfpVaAU0FIJuixE9pJKvcn Yr6UNfLJDotZr5c51St6qagqgOwBGBdfsKWsJAbTvn26jv4w12zK52I0kcguE/xF +fvwYMmVO0M= =jyNC -----END PGP SIGNATURE----- --Sig_/g3.94vV3oovgjsBx/2YeQpu--