From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <hobywank@hotmail.com>
Received: from DUB004-OMC2S4.hotmail.com (dub004-omc2s4.hotmail.com
 [157.55.1.143]) by dpdk.org (Postfix) with ESMTP id 554655A63
 for <dev@dpdk.org>; Fri,  8 May 2015 12:26:39 +0200 (CEST)
Received: from DUB131-W15 ([157.55.1.138]) by DUB004-OMC2S4.hotmail.com over
 TLS secured channel with Microsoft SMTPSVC(7.5.7601.22751); 
 Fri, 8 May 2015 03:26:39 -0700
X-TMN: [8MTOcccxMBdii+W24YfWy0SYgt8bHeNX]
X-Originating-Email: [hobywank@hotmail.com]
Message-ID: <DUB131-W15C766C6F0FAE75C5F5E12C0DE0@phx.gbl>
From: Hobywan Kenoby <hobywank@hotmail.com>
To: "Wiles, Keith" <keith.wiles@intel.com>, Avi Kivity
 <avi@cloudius-systems.com>, "O'Driscoll, Tim" <tim.o'driscoll@intel.com>
Date: Fri, 8 May 2015 12:26:39 +0200
Importance: Normal
In-Reply-To: <D1718134.1F52C%keith.wiles@intel.com>
References: <26FA93C7ED1EAA44AB77D62FBE1D27BA54D1A917@IRSMSX102.ger.corp.intel.com>,
 <26FA93C7ED1EAA44AB77D62FBE1D27BA54D29B55@IRSMSX102.ger.corp.intel.com>,
 <CAF950W+pf9iHXm4=bLf3YPCimvEQ-5fQ6o0b51AKVUW5so1yQg@mail.gmail.com>,
 <D170C88B.1F414%keith.wiles@intel.com>,
 <554B85D5.6010808@cloudius-systems.com>,
 <D170D5F0.1F48D%keith.wiles@intel.com>,
 <554B8D48.7010900@cloudius-systems.com>,
 <D1718134.1F52C%keith.wiles@intel.com>
MIME-Version: 1.0
X-OriginalArrivalTime: 08 May 2015 10:26:39.0038 (UTC)
 FILETIME=[77F5A5E0:01D08979]
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
X-Content-Filtered-By: Mailman/MimeDel 2.1.15
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] Beyond DPDK 2.0
X-BeenThere: dev@dpdk.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: patches and discussions about DPDK <dev.dpdk.org>
List-Unsubscribe: <http://dpdk.org/ml/options/dev>,
 <mailto:dev-request@dpdk.org?subject=unsubscribe>
List-Archive: <http://dpdk.org/ml/archives/dev/>
List-Post: <mailto:dev@dpdk.org>
List-Help: <mailto:dev-request@dpdk.org?subject=help>
List-Subscribe: <http://dpdk.org/ml/listinfo/dev>,
 <mailto:dev-request@dpdk.org?subject=subscribe>
X-List-Received-Date: Fri, 08 May 2015 10:26:39 -0000


> Sounds like you want something like libc=2C but DPDK is a system like a u=
ser
> space OS more then it is a collection of functions that are independent
> like strlen=2C strcpy=2C memcpy=2C printf or ... Some parts of DPDK are
> independent and can be used as you suggest=2C but the real performance
> sections are tied together.
>=20
> >> Regards=2C
> >> ++Keith

This is indeed quite a statement. DPDK is not just a=0A=
bunch of NIC drivers=2C but "a user space OS" (DPDK 1.0 had a baremetal=0A=
boot: why did it disappeared?).=0A=
=0A=
 =0A=
=0A=
Why Linux or Windows do not integrate DPDK concepts to=0A=
catch up performance wise? Is it something so deep like the "Big=0A=
Kernel Lock" that took so many years to get rid of?=0A=
=0A=
 =0A=
=0A=
My assumption is that all current kernels have been=0A=
built with one implicit hypothesis: the memory is much faster than cpu. Thi=
s is=0A=
the opposite today. DPDK internal structure has been adapted to the new=0A=
paradigm where the TLBs=2C the memory bandwidth are the scarce resources to=
 manage. So I=0A=
guess Linux and Windows will not be able to integrate DPDK concepts for=0A=
performance anytime soon=2C if ever...=0A=
=0A=
 =0A=
=0A=
Reading the list carefully=2C I expect disk block PMDs=0A=
(and block framework?) to come next.=0A=
=0A=
 =0A=
=0A=
Beyond DPDK 2.0: is it time to accept the fact that=0A=
DPDK community is actually paving the way to the next generation lightweigh=
t=2C=0A=
high performance=2C para-virtualized OS? Is it a DPDK task? Another project=
 ?=0A=
Should we rename DPDK to PVDK?=0A=
=0A=
 =0A=
=0A=
- HK=0A=
=0A=
  		 	   		  =