From: keith.busch@intel.com (Keith Busch)
Subject: [PATCH] nvme-multipath: implement active-active round-robin path selector
Date: Wed, 4 Apr 2018 08:30:27 -0600 [thread overview]
Message-ID: <20180404143027.GE4485@localhost.localdomain> (raw)
In-Reply-To: <e22676055aca4e8d936b7479d6301d85@skt-tnetpmx2.SKT.AD>
On Fri, Mar 30, 2018@09:04:46AM +0000, Eric H. Chang wrote:
> We internally call PCIe-retimer as HBA. It's not a real Host Bus Adapter that translates the interface from PCIe to SATA or SAS. Sorry for the confusion.
Please don't call a PCIe retimer an "HBA"! :)
While your experiment is setup to benefit from round-robin, my only
concern is it has odd performance in a real world scenario with
IO threads executing in different nodes. Christoph's proposal will
naturally utilize both paths optimally there, where round-robin will
saturate node interlinks.
Not that I'm against having the choice; your setup probably does represent
real use. But if we're going to have multiple choice, user documentation
on nvme path selectors will be useful here.
WARNING: multiple messages have this Message-ID (diff)
From: Keith Busch <keith.busch@intel.com>
To: "Eric H. Chang" <echang@sk.com>
Cc: Christoph Hellwig <hch@lst.de>, Baegjae Sung <baegjae@gmail.com>,
"axboe@fb.com" <axboe@fb.com>,
"sagi@grimberg.me" <sagi@grimberg.me>,
"linux-nvme@lists.infradead.org" <linux-nvme@lists.infradead.org>,
"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>
Subject: Re: [PATCH] nvme-multipath: implement active-active round-robin path selector
Date: Wed, 4 Apr 2018 08:30:27 -0600 [thread overview]
Message-ID: <20180404143027.GE4485@localhost.localdomain> (raw)
In-Reply-To: <e22676055aca4e8d936b7479d6301d85@skt-tnetpmx2.SKT.AD>
On Fri, Mar 30, 2018 at 09:04:46AM +0000, Eric H. Chang wrote:
> We internally call PCIe-retimer as HBA. It's not a real Host Bus Adapter that translates the interface from PCIe to SATA or SAS. Sorry for the confusion.
Please don't call a PCIe retimer an "HBA"! :)
While your experiment is setup to benefit from round-robin, my only
concern is it has odd performance in a real world scenario with
IO threads executing in different nodes. Christoph's proposal will
naturally utilize both paths optimally there, where round-robin will
saturate node interlinks.
Not that I'm against having the choice; your setup probably does represent
real use. But if we're going to have multiple choice, user documentation
on nvme path selectors will be useful here.
next prev parent reply other threads:[~2018-04-04 14:30 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-27 4:38 [PATCH] nvme-multipath: implement active-active round-robin path selector Baegjae Sung
2018-03-27 4:38 ` Baegjae Sung
2018-03-28 8:06 ` Christoph Hellwig
2018-03-28 8:06 ` Christoph Hellwig
2018-03-28 19:47 ` Keith Busch
2018-03-28 19:47 ` Keith Busch
2018-03-29 8:56 ` Christoph Hellwig
2018-03-29 8:56 ` Christoph Hellwig
2018-03-30 4:57 ` Baegjae Sung
2018-03-30 4:57 ` Baegjae Sung
2018-03-30 7:06 ` Christoph Hellwig
2018-03-30 7:06 ` Christoph Hellwig
2018-03-30 9:04 ` Eric H. Chang
2018-04-04 14:30 ` Keith Busch [this message]
2018-04-04 14:30 ` Keith Busch
2018-04-05 10:11 ` Eric H. Chang
2018-04-04 12:36 ` Sagi Grimberg
2018-04-04 12:36 ` Sagi Grimberg
2018-04-04 12:39 ` Sagi Grimberg
2018-04-04 12:39 ` Sagi Grimberg
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20180404143027.GE4485@localhost.localdomain \
--to=keith.busch@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.