Return-Path: Received: from userp1040.oracle.com ([156.151.31.81]:34434 "EHLO userp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751146AbcCGPlk (ORCPT ); Mon, 7 Mar 2016 10:41:40 -0500 To: Shaun Tancheff Cc: Boaz Harrosh , "Martin K. Petersen" , Andreas Dilger , Jens Axboe , linux-fsdevel@vger.kernel.org, linux-block@vger.kernel.org, calvinowens@fb.com, hch@lst.de Subject: Re: [PATCH 0/11] Update version of write stream ID patchset From: "Martin K. Petersen" References: <1457107853-8689-1-git-send-email-axboe@fb.com> <56AF7A2D-DF4F-4973-B4F6-6F515A8C500C@dilger.ca> <56DC55E4.8070205@plexistor.com> Date: Mon, 07 Mar 2016 10:41:21 -0500 In-Reply-To: (Shaun Tancheff's message of "Sun, 6 Mar 2016 14:51:53 -0600") Message-ID: MIME-Version: 1.0 Content-Type: text/plain Sender: linux-fsdevel-owner@vger.kernel.org List-ID: >>>>> "Shaun" == Shaun Tancheff writes: Shaun> This presupposes that the device itself is attached to a single Shaun> machine and the kernel is in full control of the device. Currently the NVMe TWG is leaning towards having each "I-T nexus" have dedicated stream resources. So if your controller supports 16 streams, in a 4-way multipath setup each host would have a cap of 4 concurrent streams. -- Martin K. Petersen Oracle Linux Engineering