Re: [dtn-users] Delay in receiving files due to link down

Carlo Caini <carlo.caini@unibo.it> Tue, 14 July 2015 07:52 UTC

Return-Path: <prvs=2637152654=carlo.caini@unibo.it>
X-Original-To: dtn-users@ietfa.amsl.com
Delivered-To: dtn-users@ietfa.amsl.com
Received: from localhost (ietfa.amsl.com [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 414AA1A902E for <dtn-users@ietfa.amsl.com>; Tue, 14 Jul 2015 00:52:06 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: 3.917
X-Spam-Level: ***
X-Spam-Status: No, score=3.917 tagged_above=-999 required=5 tests=[BAYES_20=-0.001, HELO_EQ_IT=0.635, HOST_EQ_IT=1.245, J_CHICKENPOX_54=0.6, RCVD_IN_BRBL_LASTEXT=1.449, SPF_PASS=-0.001, T_RP_MATCHES_RCVD=-0.01] autolearn=no
Received: from mail.ietf.org ([4.31.198.44]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id DUnnmzaigX7S for <dtn-users@ietfa.amsl.com>; Tue, 14 Jul 2015 00:52:04 -0700 (PDT)
Received: from mx01.unibo.it (mx01.unibo.it [137.204.24.54]) by ietfa.amsl.com (Postfix) with ESMTP id 14D1E1A902C for <dtn-users@irtf.org>; Tue, 14 Jul 2015 00:52:03 -0700 (PDT)
X-AuditID: 89cc1836-f79676d000004656-7e-55a4bfa2d350
Received: from mail.unibo.it (e10-hc4-dr.unibo.loc [10.11.1.42]) by mx01.unibo.it (UNIBO) with SMTP id 47.F3.18006.2AFB4A55; Tue, 14 Jul 2015 09:52:02 +0200 (CEST)
Received: from ccaini-PC.unibo.it (137.204.143.2) by mail.unibo.it (10.11.1.42) with Microsoft SMTP Server (TLS) id 14.3.224.2; Tue, 14 Jul 2015 09:52:01 +0200
X-Mailer: QUALCOMM Windows Eudora Version 7.1.0.9
Date: Tue, 14 Jul 2015 09:51:56 +0200
To: Rucha Vaidya <rucha.vaidya21@gmail.com>, dtn-users@irtf.org
From: Carlo Caini <carlo.caini@unibo.it>
In-Reply-To: <CAN=tgUw6xUVJFet+wAKrUo66POLpJrvzsGAKpoPde2pf8PtBpw@mail.g mail.com>
References: <CAN=tgUw6xUVJFet+wAKrUo66POLpJrvzsGAKpoPde2pf8PtBpw@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format="flowed"
Message-ID: <9f967b6f-c509-411d-bb1e-aa3e150a112b@E10-HC4-DR.personale.dir.unibo.it>
X-Originating-IP: [137.204.143.2]
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFlrKIsWRmVeSWpSXmKPExsXCxc2opbto/5JQg82dchZX775isjjU187i wOSxc9Zddo/JGw+zBTBFcdskJZaUBWem5+nbJXBnXP54m6XgjXjFl50/2BsY5wp3MXJySAiY SNxce4QRwhaTuHBvPVsXIxeHkMBSRontv76wQDiLGSWuHvsElOEAqjKU2L6ABaSBRUBV4uPj bUwgtoiAo8Sf119ZQWw2AQ2Jg7cus4OUCws4Sfy4EAcS5hQIkVi8dRUbiC0kECCxbd4rZhCb V0BQ4uTMJywg5cwC1hLrH2VBhMMklj/8wwxRriix7tAUqDMVJXau2sg+gVFgFpLuWQjdCxiZ VjGKFeem6xYnJ+YZ6iUX65XmZSbl6+XkJ29iBAZf5xkJsx2Mq867HWKU5GBSEuX9s3BJqBBf Un5KZUZicUZ8UWlOavEhRhkODiUJXuF9QDnBotT01Iq0zBxgHMCkmTg4DzFKcHBJiRSn5qWk FiWWlmTEg0I8vhgY5CApHiURXm+Qdt7igsRcoChE6ylGRSlxXneQhABIIqM0D26sEth5/Uyv GMU5GJWEeSVBqniAcQ/X/QpoMBPQ4Llii0AGlyQipKQaGBlSIy689NnF/kJP4vKaMv5nUTPv Lwm6OP2us0CR7Re9eG6bjeV97+6Yxayq65rbFv6lsOlw7fq20zvmC7JZNpw7sKHRvS/4U9zL 93N6VmXPmv/haXLJbJUOjp4DF9qXpEpVPS/gYLd5fvrQ0nNTdP8FifaYH469yPBeS+xuYXy5 gUPM1oPZR5VYijMSDbWYi4oTAaFnmM64AgAA
Archived-At: <http://mailarchive.ietf.org/arch/msg/dtn-users/amIYGoYWgdbaDzY0LFqbTf3I_C4>
Subject: Re: [dtn-users] Delay in receiving files due to link down
X-BeenThere: dtn-users@irtf.org
X-Mailman-Version: 2.1.15
Precedence: list
List-Id: "The Delay-Tolerant Networking Research Group \(DTNRG\) - Users." <dtn-users.irtf.org>
List-Unsubscribe: <https://www.irtf.org/mailman/options/dtn-users>, <mailto:dtn-users-request@irtf.org?subject=unsubscribe>
List-Archive: <https://mailarchive.ietf.org/arch/browse/dtn-users/>
List-Post: <mailto:dtn-users@irtf.org>
List-Help: <mailto:dtn-users-request@irtf.org?subject=help>
List-Subscribe: <https://www.irtf.org/mailman/listinfo/dtn-users>, <mailto:dtn-users-request@irtf.org?subject=subscribe>
X-List-Received-Date: Tue, 14 Jul 2015 07:52:06 -0000

Dear Rucha,
    you can find some insights on DTN2 channel probing here:

C. Caini, R. Firrincieli, M. Livini, "DTN Bundle Layer over TCP: 
Retransmission Algorithms in the Presence of Channel Disruptions", in 
Journal of Communications (JCM), Academy Publisher, special issue on 
Delay Tolerant Networks, Architecture, and Applications, Vol. 5, N. 
2, pp. 106-116, February, 2010.

In brief, a disruption slow down bundle delivery because of:
1) the channel is not available during the disruption, as obvious;
2) the availability of the channel is probed at given (exponentially 
increasing until a ceiling is reached) intervals. Therefore there is 
a delay between channel availability and recognition at bundle layer 
of this availability (on average a half of the retry interval); see 
the response by Vangelis on how to reduce this;
3) if you use TCP at convergence layer, a brand new TCP connection is 
open when a probing attempt is successful; this connection has the 
CWND set to the minumum and it may take some time, in the presence of 
long RTTs, to increase enough to exploit the bandwidth available at 
physical layer; this effect, as said, is strongly related to your 
RTT. It is significant for GEO satellite channels (RTT= about 600ms), 
much less for terrestrial connections.

Two other considerations:
1) scheduled links are not actually implemented in DTN2 (there is 
only a stub); they could be useful if you knew the contacts in advance;
2) to the better of my knowledge (maybe I am wrong) priorities can be 
set but standard static routers does not enforce them in DTN2. 
Anyway, priorities would not have any effect if all the traffic had 
the same priority, as likely in your case.

Yours,
     Carlo



At 20:46 12/07/2015, Rucha Vaidya wrote:
>Hello,
>I have installed dtn-2.9.0 and have deployed it on a network where I 
>am checking it for connections which last for a very short time and 
>then get disconnected. So I am currently using dtncpd and dtncp to 
>transfer files from one node to another. However when the link is 
>put down for some time and if I send a lot of files(even as less as 
>10) and then the link is put up it takes a lot of time for the files 
>to get received by the dtncpd. As per our observations, the time 
>taken was dependent on the duration for which the link was down and 
>the number of files sent. Is there any configuration through which I 
>can change this. Can it be made to tranfer atleast some files rather 
>than nothing at all, because the application won't be possible if 
>transfer doesn't happen in short duration.
>Along the lines I read: the <https://tools.ietf.org/html/rfc5050>RFC 
>5050 Bundle Protocol specification includes "bulk", "normal", and 
>"expedited" markings. Does dtncpd have this which might help me in 
>solving the above problem?
>
>Thanks
>[]
>Rucha
>_______________________________________________
>dtn-users mailing list
>dtn-users@irtf.org
>https://www.irtf.org/mailman/listinfo/dtn-users