From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <m.heiserer@proxmox.com>
Received: from firstgate.proxmox.com (firstgate.proxmox.com [212.224.123.68])
 (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)
 key-exchange X25519 server-signature RSA-PSS (2048 bits))
 (No client certificate requested)
 by lists.proxmox.com (Postfix) with ESMTPS id E4985702A6
 for <pve-devel@lists.proxmox.com>; Mon, 13 Jun 2022 11:59:56 +0200 (CEST)
Received: from firstgate.proxmox.com (localhost [127.0.0.1])
 by firstgate.proxmox.com (Proxmox) with ESMTP id D227D9776
 for <pve-devel@lists.proxmox.com>; Mon, 13 Jun 2022 11:59:26 +0200 (CEST)
Received: from proxmox-new.maurer-it.com (proxmox-new.maurer-it.com
 [94.136.29.106])
 (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits)
 key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256)
 (No client certificate requested)
 by firstgate.proxmox.com (Proxmox) with ESMTPS id 1AFA4976B
 for <pve-devel@lists.proxmox.com>; Mon, 13 Jun 2022 11:59:26 +0200 (CEST)
Received: from proxmox-new.maurer-it.com (localhost.localdomain [127.0.0.1])
 by proxmox-new.maurer-it.com (Proxmox) with ESMTP id 4BDAF42DD2
 for <pve-devel@lists.proxmox.com>; Mon, 13 Jun 2022 11:59:25 +0200 (CEST)
Message-ID: <fb69c72f-e83c-b42e-225b-32460d4b1284@proxmox.com>
Date: Mon, 13 Jun 2022 11:59:24 +0200
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101
 Thunderbird/91.10.0
Content-Language: en-US
To: Fabian Ebner <f.ebner@proxmox.com>, pve-devel@lists.proxmox.com
References: <20220527114810.510501-1-m.heiserer@proxmox.com>
 <20220527114810.510501-3-m.heiserer@proxmox.com>
 <acf8204b-881c-7a63-49b9-a9b7321d9b6c@proxmox.com>
From: Matthias Heiserer <m.heiserer@proxmox.com>
In-Reply-To: <acf8204b-881c-7a63-49b9-a9b7321d9b6c@proxmox.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
X-SPAM-LEVEL: Spam detection results:  0
 AWL 0.674 Adjusted score from AWL reputation of From: address
 BAYES_00                 -1.9 Bayes spam probability is 0 to 1%
 KAM_DMARC_STATUS 0.01 Test Rule for DKIM or SPF Failure with Strict Alignment
 NICE_REPLY_A           -1.732 Looks like a legit reply (A)
 SPF_HELO_NONE           0.001 SPF: HELO does not publish an SPF Record
 SPF_PASS               -0.001 SPF: sender matches SPF record
 T_SCC_BODY_TEXT_LINE    -0.01 -
Subject: Re: [pve-devel] [PATCH v4 manager] HDEdit: check iothread by
 default and move it from advanced section
X-BeenThere: pve-devel@lists.proxmox.com
X-Mailman-Version: 2.1.29
Precedence: list
List-Id: Proxmox VE development discussion <pve-devel.lists.proxmox.com>
List-Unsubscribe: <https://lists.proxmox.com/cgi-bin/mailman/options/pve-devel>, 
 <mailto:pve-devel-request@lists.proxmox.com?subject=unsubscribe>
List-Archive: <http://lists.proxmox.com/pipermail/pve-devel/>
List-Post: <mailto:pve-devel@lists.proxmox.com>
List-Help: <mailto:pve-devel-request@lists.proxmox.com?subject=help>
List-Subscribe: <https://lists.proxmox.com/cgi-bin/mailman/listinfo/pve-devel>, 
 <mailto:pve-devel-request@lists.proxmox.com?subject=subscribe>
X-List-Received-Date: Mon, 13 Jun 2022 09:59:57 -0000

On 10.06.2022 11:20, Fabian Ebner wrote:
> Am 27.05.22 um 13:48 schrieb Matthias Heiserer:
>> Existing disks are not changed by this.
>> Especially in benchmarks, iothreads significantly improve IO performance.
>>
>> Signed-off-by: Matthias Heiserer <m.heiserer@proxmox.com>
> 
> Reviewed-by: Fabian Ebner <f.ebner@proxmox.com>
> 
>> ---
>>
>> Changes from v3:
>> * remove automatically switching to/from SCSI single when iothread is (un)checked
>> * iothread will be initially set on
>> * changing the controller to something other than SCSI single and then back to
>>    SCSI single will enable iothread for all SCSI disks.
> 
> I like this version much more. We could still add validation/warning to
> the checkbox, but IMHO not a big deal with the new default and improved
> warning upon VM start. Just wondering if we should enable iothread by
> default for virtio disks too, or is the performance improvement not as
> relevant there?
Okay great, then I won't change the UI in v5.

Just did a very quick test and on my machine it seems to make read 
benchmarks faster, but write is slightly slower. Will enable it in v5.

fio --ioengine=libaio --direct=1 --sync=1 --rw=read --bs=4k --numjobs=1 
--iodepth=1 --runtime=60 --time_based --name seq_read --filename=/dev/vdb

With iothread:
read: IOPS=27.1k, BW=106MiB/s
write: IOPS=707, BW=2831KiB/s

Without iothread:
read: IOPS=13.4k, BW=52.5MiB/s
write: IOPS=787, BW=3150KiB/s