All of lore.kernel.org
 help / color / mirror / Atom feed
From: Alexey Lyashkov <alexey.lyashkov@gmail.com>
To: NeilBrown <neilb@suse.com>
Cc: "Dilger, Andreas" <andreas.dilger@intel.com>,
	Andreas Dilger <adilger@dilger.ca>,
	"devel@driverdev.osuosl.org" <devel@driverdev.osuosl.org>,
	"Drokin, Oleg" <oleg.drokin@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Christoph Hellwig <hch@infradead.org>,
	"selinux@tycho.nsa.gov" <selinux@tycho.nsa.gov>,
	fsdevel <linux-fsdevel@vger.kernel.org>,
	lustre-devel <lustre-devel@lists.lustre.org>
Subject: Re: [lustre-devel] [PATCH] staging: lustre: delete the filesystem from the tree.
Date: Mon, 4 Jun 2018 06:59:45 +0300	[thread overview]
Message-ID: <FE290581-D17D-4CEC-9AE2-B79DCBBD0823@gmail.com> (raw)
In-Reply-To: <87h8mjp5o0.fsf@notabene.neil.brown.name>


> 4 июня 2018 г., в 6:54, NeilBrown <neilb@suse.com> написал(а):
> 
> On Sun, Jun 03 2018, Dilger, Andreas wrote:
> 
>> On Jun 1, 2018, at 17:19, NeilBrown <neilb@suse.com> wrote:
>>> 
>>> On Fri, Jun 01 2018, Doug Oucharek wrote:
>>> 
>>>> Would it makes sense to land LNet and LNDs on their own first?  Get
>>>> the networking house in order first before layering on the file
>>>> system?
>>> 
>>> I'd like to turn that question on it's head:
>>> Do we need LNet and LNDs?  What value do they provide?
>>> (this is a genuine question, not being sarcastic).
>>> 
>>> It is a while since I tried to understand LNet, and then it was a
>>> fairly superficial look, but I think it is an abstraction layer
>>> that provides packet-based send/receive with some numa-awareness
>>> and routing functionality.  It sits over sockets (TCP) and IB and
>>> provides a uniform interface.
>> 
>> LNet is originally based on a high-performance networking stack called
>> Portals (v3, http://www.cs.sandia.gov/Portals/), with additions for LNet
>> routing to allow cross-network bridging.
>> 
>> A critical part of LNet is that it is for RDMA and not packet-based
>> messages.  Everything in Lustre is structured around RDMA.  Of course,
>> RDMA is not possible with TCP

To be clear. Soft IB (aka Soft RoCE) driver is part of OFED stack from 4.8(or 4.9).
So RDMA API now is possible with TCP networks.


Alex

WARNING: multiple messages have this Message-ID (diff)
From: Alexey Lyashkov <alexey.lyashkov@gmail.com>
To: NeilBrown <neilb@suse.com>
Cc: "Dilger, Andreas" <andreas.dilger@intel.com>,
	Andreas Dilger <adilger@dilger.ca>,
	"devel@driverdev.osuosl.org" <devel@driverdev.osuosl.org>,
	"Drokin, Oleg" <oleg.drokin@intel.com>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
	Christoph Hellwig <hch@infradead.org>,
	"selinux@tycho.nsa.gov" <selinux@tycho.nsa.gov>,
	fsdevel <linux-fsdevel@vger.kernel.org>,
	lustre-devel <lustre-devel@lists.lustre.org>
Subject: [lustre-devel] [PATCH] staging: lustre: delete the filesystem from the tree.
Date: Mon, 4 Jun 2018 06:59:45 +0300	[thread overview]
Message-ID: <FE290581-D17D-4CEC-9AE2-B79DCBBD0823@gmail.com> (raw)
In-Reply-To: <87h8mjp5o0.fsf@notabene.neil.brown.name>


> 4 ???? 2018 ?., ? 6:54, NeilBrown <neilb@suse.com> ???????(?):
> 
> On Sun, Jun 03 2018, Dilger, Andreas wrote:
> 
>> On Jun 1, 2018, at 17:19, NeilBrown <neilb@suse.com> wrote:
>>> 
>>> On Fri, Jun 01 2018, Doug Oucharek wrote:
>>> 
>>>> Would it makes sense to land LNet and LNDs on their own first?  Get
>>>> the networking house in order first before layering on the file
>>>> system?
>>> 
>>> I'd like to turn that question on it's head:
>>> Do we need LNet and LNDs?  What value do they provide?
>>> (this is a genuine question, not being sarcastic).
>>> 
>>> It is a while since I tried to understand LNet, and then it was a
>>> fairly superficial look, but I think it is an abstraction layer
>>> that provides packet-based send/receive with some numa-awareness
>>> and routing functionality.  It sits over sockets (TCP) and IB and
>>> provides a uniform interface.
>> 
>> LNet is originally based on a high-performance networking stack called
>> Portals (v3, http://www.cs.sandia.gov/Portals/), with additions for LNet
>> routing to allow cross-network bridging.
>> 
>> A critical part of LNet is that it is for RDMA and not packet-based
>> messages.  Everything in Lustre is structured around RDMA.  Of course,
>> RDMA is not possible with TCP

To be clear. Soft IB (aka Soft RoCE) driver is part of OFED stack from 4.8(or 4.9).
So RDMA API now is possible with TCP networks.


Alex

  reply	other threads:[~2018-06-04  3:59 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-06-01  9:11 [PATCH] staging: lustre: delete the filesystem from the tree Greg Kroah-Hartman
2018-06-01  9:11 ` [lustre-devel] " Greg Kroah-Hartman
2018-06-01 11:41 ` Christoph Hellwig
2018-06-01 11:41   ` [lustre-devel] " Christoph Hellwig
2018-06-01 18:20   ` Andreas Dilger
2018-06-01 18:20     ` [lustre-devel] " Andreas Dilger
2018-06-01 18:25     ` Doug Oucharek
2018-06-01 18:25       ` Doug Oucharek
2018-06-01 23:19       ` NeilBrown
2018-06-01 23:19         ` NeilBrown
2018-06-03 20:34         ` Dilger, Andreas
2018-06-03 20:34           ` Dilger, Andreas
2018-06-04  3:54           ` NeilBrown
2018-06-04  3:54             ` NeilBrown
2018-06-04  3:59             ` Alexey Lyashkov [this message]
2018-06-04  3:59               ` Alexey Lyashkov
2018-06-04  4:15               ` Andreas Dilger
2018-06-04  4:15                 ` Andreas Dilger
2018-06-04  4:17                 ` Alexey Lyashkov
2018-06-04  4:17                   ` Alexey Lyashkov
2018-06-01 18:30 ` Andreas Dilger
2018-06-01 18:30   ` [lustre-devel] " Andreas Dilger
2018-06-01 18:41   ` Oleg Drokin
2018-06-01 18:41     ` [lustre-devel] " Oleg Drokin
2018-06-01 19:08   ` Greg Kroah-Hartman
2018-06-01 19:08     ` [lustre-devel] " Greg Kroah-Hartman
2018-06-04  7:09     ` Christoph Hellwig
2018-06-04  7:09       ` [lustre-devel] " Christoph Hellwig
2018-06-04  7:14       ` Greg Kroah-Hartman
2018-06-04  7:14         ` [lustre-devel] " Greg Kroah-Hartman
2018-06-02  0:28 ` NeilBrown
2018-06-02  0:28   ` NeilBrown

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=FE290581-D17D-4CEC-9AE2-B79DCBBD0823@gmail.com \
    --to=alexey.lyashkov@gmail.com \
    --cc=adilger@dilger.ca \
    --cc=andreas.dilger@intel.com \
    --cc=devel@driverdev.osuosl.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=hch@infradead.org \
    --cc=linux-fsdevel@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lustre-devel@lists.lustre.org \
    --cc=neilb@suse.com \
    --cc=oleg.drokin@intel.com \
    --cc=selinux@tycho.nsa.gov \
    /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.