Re: [decade] [ppsp] Object naming in -req and -arch

Songhaibin <haibin.song@huawei.com> Thu, 19 July 2012 02:37 UTC

From: Songhaibin <haibin.song@huawei.com>
To: "arno@cs.vu.nl" <arno@cs.vu.nl>, Peng Zhang <pzhang.thu@gmail.com>
Thread-Topic: [decade] [ppsp] Object naming in -req and -arch
Thread-Index: AQHNYGz9SgP+UqsLmUq+tORnpn0WHZcmRtaAgAmX1jA=
Date: Thu, 19 Jul 2012 02:35:29 +0000
Message-ID: <E33E01DFD5BEA24B9F3F18671078951F23AF0FE9@szxeml534-mbx.china.huawei.com>
References: <20120710162606039401143@chinamobile.com> <2039343B-5F6B-4777-864E-B4F00B5A258E@gmail.com> <4FFE6D67.80705@cs.vu.nl> <D282B585-33DD-4A0D-8CD3-9CF525C56446@gmail.com> <4FFFCA6D.7060306@cs.vu.nl>
In-Reply-To: <4FFFCA6D.7060306@cs.vu.nl>
Accept-Language: en-US, zh-CN
Content-Language: zh-CN
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
Cc: ppsp <ppsp@ietf.org>, decade <decade@ietf.org>
Subject: Re: [decade] [ppsp] Object naming in -req and -arch
Precedence: list

I know little about Merkle Hash Tree. So I searched and learned some from the Internet. My feeling is MHT is useful for pre-stored objects. But it is less useful for live contents. IMO, the co-relation of the objects besides naming the object itself adds extra complexity in DECADE environment. Or is there any good reason to use it?

Just my 2 cents,
-Haibin

> -----Original Message-----
> From: decade-bounces@ietf.org [mailto:decade-bounces@ietf.org] On Behalf Of
> Arno Bakker
> Sent: Friday, July 13, 2012 3:13 PM
> To: Peng Zhang
> Cc: ppsp; decade
> Subject: Re: [decade] [ppsp] Object naming in -req and -arch
> 
> On 12/07/2012 22:28, Peng Zhang wrote:
> >
> > On Jul 12, 2012, at 2:23 AM, Arno Bakker wrote:
> >
> >> The gains of using MHT depend on the chunk size. For PPSP we prefer
> >> chunks of 1K that fit in an UDP packet carried over Ethernet. In
> >> that case, for a 4 GB file, there are 4 M chunks, resulting in 80
> >> MB of leaf hashes when SHA1 is used. Transferring that beforehand
> >> as in BitTorrent definitely increases latency ;o)
> > Yes, if the chunk size is only 1KB, and each chunk is verified
> > individually, we cannot afford to send all hashes beforehand. While
> > in the worst case without optimization, almost 2*80M = 160M hashes
> > needs to be sent to the receiver, will that be a large overhead
> > compared to 4G? Do we really need such a small chunk size? Maybe I
> > miss some previous discussion on this.
> 
> Hi
> 
> For PPSP we want to use UDP as we don't need the in-order and
> reliability features of TCP, and want flexibility to use differnet
> congestion control algorithms and handle NATs. With Ethernet as the
> dominant MAC layer at present and an unreliable transport we don't want
> datagrams to exceed the Ethernet MTU, otherwise the chance of losing a
> datagram increases (an UDP packet taking N IP packets will not be
> delivered when only 1 IP packet is lost). Hence, we use chunks of ~1K.
> 
> A good practice in P2P networks is to not forward data you have not
> verified. So to forward the 1K chunks directly we need to be able to
> verify their integrity at this granularity, enter Merkle Hash Trees.
> We think the resulting overhead due to the size of the tree is
> acceptable, as it is easy to optimize the number of hashes transmitted
> in our use cases.
> 
> CU,
>      Arno

[decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Stephen Farrell
[decade] 答复: Object naming in -req and -arch Wangdanhua
[decade] 回复: 答复: Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Stephen Farrell
Re: [decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Stephen Farrell
Re: [decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Stephen Farrell
Re: [decade] Object naming in -req and -arch Y. Richard Yang
[decade] 答复: 答复: Object naming in -req and -arch Wangdanhua
Re: [decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Stephen Farrell
Re: [decade] Object naming in -req and -arch Zhang Peng
Re: [decade] Object naming in -req and -arch Dirk Kutscher
Re: [decade] Object naming in -req and -arch Peng Zhang
Re: [decade] Object naming in -req and -arch Y. Richard Yang
[decade] Fw: Re: [ppsp] Object naming in -req and… zhangyunfei
Re: [decade] [ppsp] Object naming in -req and -ar… Peng Zhang
Re: [decade] [ppsp] Object naming in -req and -ar… Arno Bakker
Re: [decade] [ppsp] Object naming in -req and -ar… Peng Zhang
Re: [decade] [ppsp] Object naming in -req and -ar… Arno Bakker
Re: [decade] [ppsp] Object naming in -req and -ar… Songhaibin