Internet DRAFT - draft-shi-dtn-amcud
draft-shi-dtn-amcud
Delay-Tolerant Networking Research Group Wenfeng Shi
Internet Draft Qi Xu
Intended status: Experimental Bohao Feng
Expires: December 24, 2018 Huachun Zhou
Beijing Jiaotong University
June 25, 2018
A Mechanism Coping with Unexpected Disruption in Space Delay
Tolerant Networks
draft-shi-dtn-amcud-06.txt
Status of this Memo
This Internet-Draft is submitted in full conformance with the
provisions of BCP 78 and BCP 79.
This document may contain material from IETF Documents or IETF
Contributions published or made publicly available before November
10, 2008. The person(s) controlling the copyright in some of this
material may not have granted the IETF Trust the right to allow
modifications of such material outside the IETF Standards Process.
Without obtaining an adequate license from the person(s) controlling
the copyright in such materials, this document may not be modified
outside the IETF Standards Process, and derivative works of it may
not be created outside the IETF Standards Process, except to format
it for publication as an RFC or to translate it into languages other
than English.
Internet-Drafts are working documents of the Internet Engineering
Task Force (IETF), its areas, and its working groups. Note that
other groups may also distribute working documents as Internet-
Drafts.
Internet-Drafts are draft documents valid for a maximum of six
months and may be updated, replaced, or obsoleted by other documents
at any time. It is inappropriate to use Internet-Drafts as
reference material or to cite them other than as "work in progress."
The list of current Internet-Drafts can be accessed at
http://www.ietf.org/ietf/1id-abstracts.txt
The list of Internet-Draft Shadow Directories can be accessed at
http://www.ietf.org/shadow.html
This Internet-Draft will expire on December 24, 2018.
Shi, et al. Expires December 24, 2018 [Page 1]
Internet-Draft amcud June 2018
Copyright Notice
Copyright (c) 2018 IETF Trust and the persons identified as the
document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal
Provisions Relating to IETF Documents
(http://trustee.ietf.org/license-info) in effect on the date of
publication of this document. Please review these documents
carefully, as they describe your rights and restrictions with
respect to this document. Code Components extracted from this
document must include Simplified BSD License text as described in
Section 4.e of the Trust Legal Provisions and are provided without
warranty as described in the Simplified BSD License.
Abstract
This document proposes a coping mechanism used to deal with the
unpredictable disruption problem and congestion control problem in
Space Delay Tolerant Networks (DTN) [RFC4838]. Since Licklider
Transmission Protocol (LTP) [RFC5326] provides retransmission-based
reliability for bundles, several times of retransmissions can be
seen as a failure occurred over links. The proposed mechanism is
used to direct the following packets to other nodes as soon as the
selected path is detected as disruption or congestion and probes the
availability of the links which has disrupted unexpectedly.
Table of Contents
1. Introduction ................................................ 2
2. Conventions used in this document............................ 3
3. The coping mechanism......................................... 3
4. Security Considerations...................................... 6
5. IANA Considerations ......................................... 7
6. References .................................................. 7
1. Introduction
Since the moving trajectory of nodes is scheduled in the space
network, it's possible to have a prior knowledge of contact
information between any nodes. Consequently, routing algorithms such
as Contact Graph Routing (CGR) [CGR] can calculate a delivery path
from the source to destination hop by hop based on the connectivity
relationship, propagation delay, data rate, etc.
Shi, et al. Expires December 24, 2018 [Page 2]
Internet-Draft amcud June 2018
However, due to the complexity of the space network, the satellite
and its associated links suffer from the electromagnetic
interference frequently and this may lead to unpredictable
disruption for a period of time. Then, the subsequent bundles sent
by the source using the initially contact information cannot be
transmitted successfully and retransmission is also occurred. As a
result, not only the timeliness of bundles cannot be guaranteed but
also limited resources of the node and link are consumed and wasted.
Thus, it is important to make a mechanism to handle the unexpected
disruption problem.
What's more, when the direct path to the destination is unreachable,
data will be stored at the intermediated nodes and this will consume
the node's storage resources. When the remaining storage space of
the contact end node is less than the contact capacity, it will
increase the risk of network congestion. However, the upstream nodes
have no chance to learn the congestion information. Routes that
calculated by the source nodes may not be the best choice. So it is
urgent to find a scheme to reflect the congestion status to the
upstream nodes.
This draft proposes a coping mechanism to deal with the contact
unexpected disruption problem and the network congestion problem.
The contact unexpected disruption coping mechanism works with
Licklider Transmission Protocol (LTP) [RFC5326] and routing
algorithms such as Contact Graph Routing (CGR) and it is used to not
only direct the following bundles to other nodes when the disruption
is occurred but also probe the availability of the disrupted links
during its claimed valid time. The congestion control mechanism
consists of contact congestion forecasting scheme and congestion-
aware data forwarding scheme. The contacts are divided into
different congestion levels according to nodes' storage resource. And
the data with different priority will be forwarded according to the
congestion level.
2. Conventions used in this document
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119 [RFC2119].
3. The coping mechanism
Since LTP provides retransmission-based reliability for bundles,
several times of retransmissions can be seen as a failure occurred
over links. Suppose CGR is used as the routing algorithm. Once the
retransmission is detected for more than two times, the contact used
Shi, et al. Expires December 24, 2018 [Page 3]
Internet-Draft amcud June 2018
in CGR is regarded as temporary corruption. Then, the node marks
this contact as temporary disrupted and recalculates the route for
subsequent bundles. Besides, a disruption advertisement for the
unavailable contact is sent to upstream nodes. When receiving the
advertisement, related nodes create disrupting contacts to prevent
the use of disrupted links indicated by the advertisement. However,
the advertisement may be useless when it arrives at some nodes whose
related contacts do not become available until the expiration of the
advertisement. Hence, a disruption advertisement group is defined to
assure the effectiveness of the contact disruption advertisement.
The group contains nodes indicated in corresponding contacts whose
"from time" are earlier than the disrupting contact's "to time".
When T seconds elapse, a probing message is sent by the node to the
destination shown in the disputed contact to check if the
connectivity has been recovered.Considering that the contact may be
disrupted caused by the damage of satellite, if the detection duration
is a fixed short value, it may incur more energy consumption. Thus, it's
necessary to set the detection duration dynamically.If the corresponding
response is received, the contact will be remarked as recovery and can
be used for the following bundles and a contact recovery advertisement
is sent to nodes belonging to the advertisement group. Otherwise, node
sends a probe message again 2T seconds later. If the corresponding
still haven t been received, the node will set the prove message 3T
seconds later. Also a maximum detection duration should be set to
guarantee the detection accuracy. In this way, the node probes the
disrupted link periodically until the contact is recovered or expired.
In the space network, the communication start time, end time and
transmission rate between two nodes is known in advance and is
configured into contact plan in CGR. Thus it is convenient to
compute the residual capacity of the contact. When the monitoring
node detects that the remaining storage capacity of the node is less
than the residual capacity of the contact whose end node is the
monitoring node, it will compute the congestion level of the contact.
If the remaining capacity of the monitoring node is less than thirty
percent of the residual capacity of the contact, the contact will be
marked as mild congested. If less than ten percent, the contact will
be marked as severe congested. If the capacity of the node is
exhausted, the contact will be marked as complete congested. When
the congestion level changed, the monitoring node will record the
new level of the contact in the contact plan and send contact
congestion advertisement to other nodes.
Shi, et al. Expires December 24, 2018 [Page 4]
Internet-Draft amcud June 2018
As soon as the other node receives the congestion advertisement, it
will update the congestion level of the corresponding contact
according to the advertisement. When calculating routes, the nodes
compute path congestion level as the highest congestion level of the
contact consisted in the path and forwarding different priority
bundles according to the path congestion level. If there exists no
congestion in the path, bundles of all priority can be forwarded in
the path. If the congestion level is mild, only urgent and standard
bundles can be forwarded. If the congestion level is severe, only
urgent bundle can be forwarded. If the congestion level is complete
congestion, all bundles should be forwarded using sub optimal path.
By this way, we can not only prevent data from been dropped when
network suffers from congestion but also leave the transmission
opportunity to high priority bundles.
+----------+
|Satellite2|
+----------+
/ | \
/ | \
/ | \
/ | \
+----------+ | +----------+ +----------+
|Satellite1| | |Satellite4|------|Satellite5|
+----------+ | +----------+ +----------+
\ | /
\ | /
\ | /
\ | /
+----------+
|Satellite3|
+----------+
Fig. 1 Example of unexpected contact disruption and congestion control.
An example is given to explain the contact disruption handling
mechanism. Assume that the contact between Satellite1 and Satellite2
is available from 1s to 300s, the contact between Stallite1 and
Satellite3 from 100s to 300s, the contact between Satellite3 and
Satellite4 from 100s to 300s, the contact between Satellite2 and
Satellite4 from 1s to 300s, the contact between Satellite2 and
Satellite3 from 1s to 300s, the contact between Satellite4 and
Satellite 5 from 400s to 500s. Either Satellite2 or Satellite3 can
Shi, et al. Expires December 24, 2018 [Page 5]
Internet-Draft amcud June 2018
be used by Satellite1 as relays to send bundles to Satellite5. At
initial, Satellite2 is selected to be used. Suppose at one time,
the link from Satellite2 to Satellite4 is disrupted. When Satellite2
detects the retransmission of bundles two times, it marks the
contact to Satellite4 as "temporary disrupted" and recalculates
routes for the subsequent bundles. Thus, those bundles will be sent to
Satellite3 and then to Satellite4 and Satellite5. In addition,
the disruption advertisement group is computed by Satellite2
containing Satellite1, Satellite3 and Satellite4. When Satellite1
receives the advertisement, it will mark the contact from Satellite2
to Satellite4 as "disrupted" and use Satellite3 as the relay.
At the same time, Satellite2 will send the probe message to
Satellite4 periodically and check if the link is recovered. If
Satellite2 receives a response, it will mark the contact as
"recovery" and send contact recovery advertisement to satellites
included in the advertisement group. If Satellite2 does not receive
a response after sending the probing messages, it will resend the
probing message again after T seconds. If Satellite2 still haven't
received the response after 2T seconds, it will resend the probing
message after 3T seconds. Assuming that the maximum detection
duration is set to 3T. If satellite2 still haven't received the
response after 3T, it will resend the probing message after 3T
seconds until the disrupted contact is recovered or expired.
Another example is also given to explain the congestion control
scheme. Assume that the storage capacity of satellite2 in figure 1
is 100Mbytes, the storage capacity of other satellites is 200Mbytes.
Assume thatsatellite1 sends one bulk bundle, one standard bundle and
one urgent bundle to satellite5 every second. We also assume that
the transmission rate is 200kbytes/s and the bundle size is 50kbytes.
Initially, Satellite2 is selected to be used. Since the contact time
between satellite2 and satellite4 is 100s, bundles will be stored at
satellite2 before the contact started. At the start of the
transmission, there exists no congestion. With the increase of data
stored at satellite2, the storage capacity decreased and when the
storage is less than thirty percent of the capacity between
satellite1 and sateliite2, satellite2 will find that the contact
between satellite1 and satellite2 is mild congested. It will send
congestion advertisement to satellite1. After satellite1 receives
the advertisement, it will mark the contact between satellite1 and
satellite2 as mild congested and using satellite3 as the relay for
bulk bundles. The standard and urgent bundle still be forwarded
using satellite2 as relay. When satellite2 detects the contact
between satellite1 and satellite2 is server congested, it will send
congestion advertisement to satellite1 and after satellite1 updates
the congestion level, it will forward bulk and standard bundle using
Shi, et al. Expires December 24, 2018 [Page 6]
Internet-Draft amcud June 2018
satellite3 as relay. When the storage capacity of satellite2
exhausted, the contact between satellite1 and satellite2 is complete
congested. satellite2 will send congestion advertisement to
satellite1. After satellite1 receives and updates the contact plan,
it will use satellite3 as relay for all bundles.
4. Security Considerations
To be done.
5. IANA Considerations
To be done.
6. References
[RFC4838] Burleigh S, Hooke A, Torgerson L, et al. RFC4838-Delay-
Tolerant Networking Architecture[J]. 2007.
[RFC5326] Ramadas M, Burleigh S, Farrell S. RFC 5326, Licklider
Transmission Protocol Specification[J]. IRTF DTN Research
Group, 2008.
[RFC5050] Burleigh, S. Bundle protocol specification. No. RFC 5050.
2007.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119, March 1997.
[I-D. burleigh-dtnrg-cgr] Burleigh S. Contact Graph Routing: draft-
burleigh-dtnrg-cgr-01, July 2010[J].
Shi, et al. Expires December 24, 2018 [Page 7]
Internet-Draft amcud June 2018
Authors' Addresses
Wenfeng Shi
Beijing Jiaotong University
Beijing, 100044, P.R. China
Email: 14111038@bjtu.edu.cn
Qi Xu
Beijing Jiaotong University
Beijing, 100044, P.R. China
Email: 15111046@bjtu.edu.cn
Bohao Feng
Beijing Jiaotong University
Beijing, 100044, P.R. China
Email: 11111021@bjtu.edu.cn
Huachun Zhou
Beijing Jiaotong University
Beijing, 100044, P.R. China
Email: hchzhou@bjtu.edu.cn
Shi, et al. Expires December 24, 2018 [Page 8]
