Internet DRAFT - draft-lee-icnrg-domainbasedrouting

draft-lee-icnrg-domainbasedrouting



Network Working Group                                          J. Lee
Internet Draft                                                   ETRI
Intended status: Informational                                 W. Lim
Expires: March 2015                                              ETRI
                                                              W. Chun
                                                                 HUFS
                                                     September 17, 2014



                   Scalable Domain-based Routing Scheme
                 draft-lee-icnrg-domainbasedrouting-02.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 March 17, 2015.




Lee                    Expires March 17, 2015                 [Page 1]

Internet-Draft      Scalable Domain-based Routing       September 2014


Copyright Notice

   Copyright (c) 2014 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.

Abstract

   Moving the focus from nodes to information objects raises scalability
   issues because the number of addressable information objects is huge
   compared to the number of nodes, so scalable routing is an important
   challenge issue of ICN. There are two types of naming scheme which
   have been proposed in the ICN literatures: hierarchical and flat. To
   guarantee clean separation of identifier from locator and scalability
   of routing, we chose flat name and designed domain-based routing. A
   significant requisite to be considered when flat name is used is an
   efficient name-resolution system (NRS). Bloomfilter-based NRS [BF
   NRS] is our proposal to this issue. Once a name is resolved into
   locator(s), discovery and delivery steps are carried out based on the
   routing scheme of locator. For scalability in routing of locator,
   network is projected into hierarchically-organized domain structure.
   A domain is a group of nodes or other domains. This composition could
   be physical or logical. Each domain has its own identifier, and the
   concatenation of domain IDs from top level domain to a certain domain
   which a node belongs to plays the role of "locator" for that node.
   Each domain has one or more domain gateways. All traffic from/to the
   domain should pass through any of domain gateways. Routing
   information based on this type of locators is exchanged among domain
   gateways by using modified link-state routing protocol which can
   suppress LSA explosion [LSR]. Thanks to this hierarchical domain
   structure, locators are highly aggregatable (which means scalable
   routing), and any node in heterogeneous network can communicate each
   other.



Table of Contents


   1. Introduction ................................................ 3
   2. Conventions used in this document ........................... 4


Lee                    Expires March 17, 2015                 [Page 2]

Internet-Draft      Scalable Domain-based Routing       September 2014


   3. Terminologies ............................................... 4
   4. Domain-based Routing Scheme ................................. 5
      4.1. Network as Hierarchically-organized Domain Structure ... 5
      4.2. Name Resolution System (NRS) ........................... 6
      4.3. Scalable Domain-Based Routing .......................... 6
         4.3.1. LSA filtering to prevent LSA flooding ............. 7
      4.4. Name based Forwarding .................................. 7
      4.5. In-network Caching ..................................... 8
      4.6. Mobility Management .................................... 8
      4.7. Inner-domain Communication ............................. 8
      4.8. Security ............................................... 8
      4.9. Transport Service ...................................... 8
      4.10. Operations ............................................ 9
         4.10.1. Presenting ....................................... 9
         4.10.2. Name Resolution & Discovery ...................... 9
         4.10.3. Delivery ........................................ 11
         4.10.4. Mobility Management ............................. 11
   5. Comparison with other ICN approaches [ICN survey] .......... 12
   6. Example scenario ........................................... 13
      6.1. Locator-based Routing : routing tables ................ 14
      6.2. Server Discovery (path discovery) ..................... 14
      6.3. Name-based Forwarding (Name-based delivery) ........... 15
   7. Security Considerations .................................... 16
   8. IANA Considerations ........................................ 16
   9. References ................................................. 16
      9.1. Informative References ................................ 16

1. Introduction

   Moving the focus from nodes to information objects raises scalability
   issues. Currently, the Internet is addressing on the order of 10^9
   nodes, whereas the number of addressable ICN objects is expected to
   be several orders of magnitude higher [ICNRG charter].  Therefore,
   scalable routing scheme is an important challenge issue of ICN.

   ICN routing locates a data object based on its name which is
   initially provided by a requestor. ICN routing may comprise 3 steps:
   a name resolution step, a discovery step, and a delivery step.
   Depending on how these steps are combined, ICN routing schemes can be
   categorized as Route-By-Name Routing (RBNR), Lookup-By-Name Routing
   (LBNR), and Hybrid Routing (HR) [2].

   To keep the advantage of separating identifier from locator we chose
   flat name scheme, which means LBNR is used as routing scheme. If LBNR
   is used a Name Resolution System (NRS) is required. An efficiency of
   NRS is a significant requisite in LBNR scheme because every name



Lee                    Expires March 17, 2015                 [Page 3]

Internet-Draft      Scalable Domain-based Routing       September 2014


   should be resolved through the NRS. Bloomfilter-based NRS [9] is our
   answer to this issue.

   Once the locator for the name is obtained, discovery and delivery
   steps may depend on the routing scheme of locator.  For scalability
   in routing scheme of locator, we projected networks into
   hierarchically-organized domain structure. A domain is a group of
   nodes of other domains, and this group could be physical or logical.
   Each domain has its own identifier, and the concatenation of these
   IDs from top level domain to a certain domain which a node belongs to
   plays the role of "locator" for that node.

   Each domain could have one or more domain gateways. All traffic
   from/to the domain should pass through any of its domain gateway.
   Routing information based on the locator is exchanged among domain
   gateways which belong to different domains by running modified link-
   state routing protocol. It suppresses LSA storm by using LSA-
   filtering rule between parent and child domain.

   Thanks to this hierarchical domain structure, locators are highly
   aggregatable, and routing operates regardless of the underlying
   network protocol.

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].

   In this document, these words will appear with that interpretation
   only when in ALL CAPS. Lower case uses of these words are not to be
   interpreted as carrying RFC-2119 significance.

   In this document, the characters ">>" preceding an indented line(s)
   indicates a compliance requirement statement using the key words
   listed above. This convention aids reviewers in quickly identifying
   or finding the explicit compliance requirements of this RFC.

3. Terminologies

   o Domain: a logical/physical group of nodes (requestor, contents
      server, or other "domain") which have similar characteristics (E.g.
      network protocol, geographical region, same type of contents,
      similar category of contents etc.).

   o Locator: locator is a concatenation of IDs of hierarchically
      organized domains.


Lee                    Expires March 17, 2015                 [Page 4]

Internet-Draft      Scalable Domain-based Routing       September 2014


   o Domain Gateway: a kind of border gateway for a domain. The domain
      gateway forwards request message or NDO data, and runs routing
      protocol carrying routing information based on the locator.

4. Domain-based Routing Scheme

4.1. Network as Hierarchically-organized Domain Structure

                      domain ID: A
        +---------------------------------------+
        |    +-----+       +-----+              |
        |    |     +-------+  1  |              |
        |    |     |       |     |              |  tier-0
        |    ++----+       +-----+              |  locator: A
        |   / |      \                          |
        |  /  |          \                      |
        | /   |             \                   |
        +/----|-----------------\---------------+
        /     |                      \
       /      |       domain ID: B      \
    +---------|----------------------------+
    |         |                            |
    |    +----++                +-----+    |
    |    |  2  +----------------+  3  |    |
    |    |     +---------+      |     |    |       tier-1
    |    +-----+         |      +-----+    |       locator: A:B
    |                  +-+----+            |
    |                  |      |            |
    |                  |      |            |
    |                  +-+----+            |
    |                 /  |     \           |
    |                /   |      \          |
    |               /    |       \         |
    +--------------/-----|--------\--------+
                  / +----+         \
                 /  | domain ID: C  \
               +----|-----------------+
               |    |                 |
               | +--+--+    +-----+   |
               | |  4  +----+  5  |   |           tier-2
               | |     |    |     |   |           locator: A:B:C
               | +-----+    +-----+   |
               |                      |
               +----------------------+
                  Figure 1 Hierarchical domain structure




Lee                    Expires March 17, 2015                 [Page 5]

Internet-Draft      Scalable Domain-based Routing       September 2014


   In our scheme network is projected as a set of domains, and each of
   which may be further decomposed into smaller domains recursively.
   Each domain has its own ID and the concatenation of consecutive
   domain IDs can play the role of "locator". Figure 1 shows an example
   of domain structure. For example, locator for node 3 is A:B, and
   locator for node 5 is A:B:C. All traffic to and from a domain should
   pass through a domain gateway. Node 2, 4 in this figure are domain
   gateways.

4.2. Name Resolution System (NRS)

   If flat name scheme is used, efficient name resolution system is very
   important. In Name resolution step a requestor queries current
   locator of a name to the NRS. For fast processing hashing technique
   (like bloom filter) could be used to implement NRS. We also proposes
   Bloom filter based NRS [BF NRS].

4.3. Scalable Domain-Based Routing

   Generally speaking, routing means the process of selecting best paths
   in a network along which to send network traffic. In other words,
   routing is the process of building network topology database and
   computing routing table based on that network topology. In our domain
   based structure, we build network topology graph by link-state
   routing protocol. However, not like existing link-state routing
   protocol, vertex of our network topology graph could be an actual
   communication entity or "a domain". This means each domain gateway
   which belongs to different domain in different tier has different
   topology graph. The higher tier a domain gateway belongs to, more
   abstracted topology graph it would have.

                      domain ID: A
          +---------------------------------------+
          |                                       |
          |    +-----+       +-----+              |
          |    |     +-------+  1  |              |
          |    |     |       |     |              |  tier-0
          |    +-----+       +-----+              |  locator: A
          |                                       |
          |                                       |
          |                                       |
          +---------------------------------------+
        Figure 2 Domain gateway 1's view for whole network topology

   o Contents server as a domain gateway




Lee                    Expires March 17, 2015                 [Page 6]

Internet-Draft      Scalable Domain-based Routing       September 2014


      Usually a contents server maintains many NDOs (i.e. entities with
      name), and therefore the contents server could be considered as
      another "domain gateway", which means it also joins routing
      procedure. But it does not need to maintain forwarding cache for
      NDOs (i.e. contents server already knows where NDOs are!).

4.3.1. LSA filtering to prevent LSA flooding

   Each domain gateway runs modified link-state routing protocol. Each
   LSA transfers "locator" information instead of IP prefix. Thus,
   locators appears in destination field of routing table.

   To suppress LSA storm each domain gateway forwards LSAs to the other
   domain gateways according to the following filtering rules:

   o LSAs in tier N domain are forwarded to all domain gateways in same
      tier domain.

   o LSAs in tier N-1 are injected to tier N gateways.

   o LSAs in tier N+1 are filtered, LSA which has aggregated locator is
      injected to tier N instead. (E.g. see Figure 6, [GW #2] delivers
      LSA which includes 0x0B:0x0C to [GW #1], locator for domain 0x0E
      (0x0B:0x0C:0x0E) is not flooded into [GW #1]).

   This results in topology reduction effect. Therefore, each domain
   gateway would have topology graph which is as reduced as possible in
   its current position (tier).

4.4. Name based Forwarding

   Under the principle of separating identifier from locator, header of
   each request message and data packet include name information only.
   Therefore, each domain gateway should know where to forward packets
   to reach the NDO. The domain gateways can get this information from
   "forwarding cache". Forwarding cache for certain name is built during
   discovery step. Forwarding cache includes following information.

   o Destination name: name of NDO, or name of requestor

   o Next-hop address: address of next-hop domain gateway. It can be
      any type of address (e.g. IPv4 address, IPv6 address, Ethernet
      address, ...)

   o Next-hop protocol id: protocol for next-hop gateway (e.g. IPv4,
      IPv6, Ethernet, ...)



Lee                    Expires March 17, 2015                 [Page 7]

Internet-Draft      Scalable Domain-based Routing       September 2014


   o Output interface

   Each domain gateway maintains routing table based on "locator". When
   the domain gateway receives packets destined for certain "name", it
   queries NRS to get current "locator" of NDO mapped with the "name".
   With this "locator" the domain gateway looks up its routing table to
   find routing table entry matching "locator". If matching one is found,
   forwarding cache for "name" is built by merging "name" and
   "forwarding information" part of the routing table entry.

   o The forwarding cache is built reactively, and maintained by
      lifetime timer.

   o Usually many forwarding cache can be shared by many request from
      many other requestor.

4.5. In-network Caching

   Under our structure, NDO can be cached on any domain gateway, or it
   could even be cached any node in the domain managed by the domain
   gateway. Once the NDO is cached on any node, the node adds its
   "locator" to the name of NDO through NRS. Afterwards, during
   discovery step cached NDO which has closer locator could be used.

4.6. Mobility Management

   Mobility support can be implemented very easily because all messages
   and data packets don't include any location-related information (e.g.
   locator). Usually if any domain gateway detects transmission failure,
   it restarts discovery step, then several forwarding caches on the
   path will be up-to date. There is almost nothing that end nodes
   should do.

4.7. Inner-domain Communication

   TBD.

4.8. Security

   TBD.

4.9. Transport Service

   TBD.





Lee                    Expires March 17, 2015                 [Page 8]

Internet-Draft      Scalable Domain-based Routing       September 2014


4.10. Operations

4.10.1. Presenting

   The first thing to do that every node (e.g. requestor, contents
   server...) attaches network is to update (or add newly) its current
   location. This is done by the domain gateway of domain which the
   nodes belongs to.

   +---------+  +-------------+  +----+  +----------------+  +--------+
   |requestor|  | domain gw   |  |NRS |  | domain gw      |  |contents|
   |         |  |for requestor|  |    |  |for contents svr|  |  svr   |
   +---------+  +-------------+  +----+  +----------------+  +--------+
        |               |           |            |               |
        |    1.0        |           |            |               |
        |-------------> |           |            |      2.0      |
        |               |   1.1     |            |<--------------|
        |               |---------->|    2.1     |               |
        |               |           |<-----------|               |
        |               |           |            |               |
        |               |           |            |               |
                   Figure 3 Presenting current location

   o 1.0 - presence message: deliver {requestor's name} to the domain
      gateway

   o 1.1 - update message: domain gateway sends update including
      {requestor's name, its locator} to NRS

   o 2.0 - presence message: deliver {name of NDO} to the domain
      gateway

   o 2.1 - update message: domain gateway sends update including {name
      of NDO, its locator} to NRS

4.10.2. Name Resolution & Discovery

   When a requestor issues request for a NDO on the network, name
   resolution and discovery procedure is carried out by domain gateways.

   During this procedure, forwarding cache entry for NDO includes
   followings:

   o NDO's ID

   o Next hop information (found in the routing table entry)



Lee                    Expires March 17, 2015                 [Page 9]

Internet-Draft      Scalable Domain-based Routing       September 2014


   Forwarding cache entry for the requestor includes followings:

   o Requestor's ID

   o Previous hop information (found in the packet header)

                                   +----+
                                   |NRS |
                                   +----+
   +---------+  +-------------+       |    +----------------+  +-------+
   |requestor|  | domain gw   |       |    | domain gw      |  |contens|
   |         |  |for requestor| .. ...| .. |for contents svr|  |  svr  |
   +---------+  +-------------+       |    +----------------+  +-------+
        |               |             |            |               |
        |    1.0        |             |            |               |
        |-------------> |    1.1      |            |               |
        |               |------------>|            |               |
        |               |    1.2      |            |               |
        |               |<------------|            |               |
        |               |             |            |               |
        |               |----+        |            |               |
        |               |    |1.3     |            |               |
        |               |<---+        |            |               |
        |               |             | 1.4        |               |
        |               |------------------------->|               |
        |               |             |            |----+          |
        |               |             |            |    |1.5       |
        |               |             |            |<---+          |
        |               |             |            |       1.6     |
        |               |             |            |-------------->|
        |               |             |            |       1.7     |
        |               |             | 1.8        |<--------------|
        |     1.9       |<-------------------------|               |
        |<--------------|             |            |               |
        |               |             |            |               |
                       Figure 4 Discovery procedure

   o 1.0 - request message: request message including {name of NDO}

   o 1.1 - lookup message: request {locator} for {name of NDO}

   o 1.2 - reply message: deliver {locator(s)} for {name of NDO}

   o 1.3 - build bi-directional forwarding cache





Lee                    Expires March 17, 2015                [Page 10]

Internet-Draft      Scalable Domain-based Routing       September 2014


   o 1.4 - forward request message: this request message includes
      {locator for requested NDO} to avoid unnecessary name resolution.
      (While this request message is being forwarded to the domain
      gateway for contents server, domain gateways on the path build bi-
      directional forwarding caches)

   o 1.5 - build bi-directional forwarding cache

   o 1.6 - deliver request message to the contents server

   o 1.7 ~ 1.9 - delivery step, deliver NDO data to the requester

4.10.3. Delivery

   After discovery step, forwarding caches are ready on path to the NDO,
   and on path to the requestor. Forwarding caches to the NDO may be
   reused by other request from other requestors who want same NDO.
   Forwarding caches to the requestor are used to deliver NDO.

   See 1.7 ~ 1.9 in Figure 4.

4.10.4. Mobility Management

   TBD.
























Lee                    Expires March 17, 2015                [Page 11]

Internet-Draft      Scalable Domain-based Routing       September 2014


5. Comparison with other ICN approaches [ICN survey]

   +-------------------------------------------------------------------+
   |                      |    PSIRP     |    NetInf   |  Domain-based |
   |                      |              |             |  Routing      |
   +-------------------------------------------------------------------|
   |Namespace structure   | flat with    |  flat with  |  flat no      |
   |                      | structure    |  structure  |  structure    |
   +-------------------------------------------------------------------|
   |Human-readable names  |    no        |     no      |     no        |
   +-------------------------------------------------------------------|
   |NDO granularity       |  Objects     |   Objects   |   Objects     |
   +-------------------------------------------------------------------|
   |Routing aggregation   |    scope     |  publisher  |   Domain      |
   |                      |  /explicit   |             |               |
   +-------------------------------------------------------------------|
   |Routing of NDO request|     NRS      | Hybrid NRS &|    NRS        |
   |                      |              | name based  |               |
   +-------------------------------------------------------------------|
   |Routing of NDO        |   Source     | Reverse     | Reverse       |
   |                      |routing using |request path/|request path   |
   |                      | BF           | direct IP   |               |
   +-------------------------------------------------------------------|
   |API                   | Publish      | Synchronous | IPlug/Dsocket |
   |                      | /Subscribe   | get         |               |
   +-------------------------------------------------------------------|
   |Transport             | IP/PSIRP     | Many        | Many          |
   |                      |              | including IP| including IP  |
   +-------------------------------------------------------------------+
               Figure 5 Comparison with other ICN approaches

   Figure 5 shows summary of different ICN approaches. PSIRP and NetInf
   are using flat name like ours, however our approach uses flat name
   without any structure for strict separation of name from location.
   Names for all of them are non-human readable. Routing is aggregated
   by the domain in our approach. All of them are using NRS for routing
   of NDO request. To route NDO to the requestor NetInf and our approach
   are using reverse request path. Regarding transport method, NetInf
   and our approach can support any protocol including IP, which means
   they support communication between nodes in heterogeneous network.








Lee                    Expires March 17, 2015                [Page 12]

Internet-Draft      Scalable Domain-based Routing       September 2014


6. Example scenario

   +-----------------------------------------------------------------+
   |                                                                 |
   |                                   (Domain Id: 0x0A, IPv4)       |
   |                         +------[GW #0]----------------------+   |
   |                         |        |                          |   |
   |                         |        |                          |   |
   |                         |        |                          |   |
   |                         |        |                          |   |
   |                         |        +--------------------------+   |
   |                         |                                       |
   |  (Domain Id:0x0B, IPv4) |                                       |
   |  +-------------------[GW #1]--------------------------------+   |
   |  |        ^^^^^^^^      |      <------------------------>   |   |
   |  |        | NRS  |----- +------< requestor A (Id: 0xFF) >   |   |
   |  |        ^^^^^^^^      |      <------------------------>   |   |
   |  |                      |                                   |   |
   |  |                      +--------------------------------+  |   |
   |  |                      |                                |  |   |
   |  |(Domain Id:0x0C, IPv6)|                                |  |   |
   |  |  +----------------[GW #2]-+ (Domain Id:0x0D, Ethernet)|  |   |
   |  |  |                   |    |    +------------------[GW #3]|   |
   |  |  |                   |    |    |   server3          |  | |   |
   |  |  |  server1          |    |    |  +========+        |  | |   |
   |  |  | +=============+   |    |    |  |        |        |  | |   |
   |  |  | | StarWars    |   |    |    |  |        +--------+  | |   |
   |  |  | | (0xEE)      |   |    |    |  |        |        |  | |   |
   |  |  | |             |   |    |    |  +========+        |  | |   |
   |  |  | |             +---+    |    |                    |  | |   |
   |  |  | +=============+   |    |    |      <------------->  | |   |
   |  |  |                   |    |    |      < requestor B >  | |   |
   |  |  |(Domain Id:0x0E, IPv4)  |    |      <  (Id: 0xAF) >  | |   |
   |  |  | +-----------[GW #4]-+  |    |      <------------->  | |   |
   |  |  | |   server2      |  |  |    |                       | |   |
   |  |  | |  +==========+  |  |  |    +-----------------------+ |   |
   |  |  | |  |          |  |  |  |                              |   |
   |  |  | |  |          +--+  |  |                              |   |
   |  |  | |  +==========+     |  |                              |   |
   |  |  | +-------------------+  |                              |   |
   |  |  +------------------------+                              |   |
   |  +----------------------------------------------------------+   |
   +-----------------------------------------------------------------+
                    Figure 6 Sample Domain Architecture





Lee                    Expires March 17, 2015                [Page 13]

Internet-Draft      Scalable Domain-based Routing       September 2014


6.1. Routing tables

   +---------------------+--------------------+--------+
   | Destination         | nexthop            | out if |
   +---------------------+--------------------+--------+
   | 0x0B:0x0C           | GW #2's address    | if1    |
   +---------------------+--------------------+--------+
   | 0x0B:0x0D           | GW #3's address    | if1    |
   +---------------------+--------------------+--------+
   | 0x0B                | -                  | if1    |
   +---------------------+--------------------+--------+
   | 0x0A                | GW #0's address    | if0    |
   +---------------------+--------------------+--------+
                 Figure 7 Example of Routing Table (GW #1)

   +---------------------+--------------------+--------+
   | Destination         | nexthop            | out if |
   +---------------------+--------------------+--------+
   | 0x0B:0x0C:0x0E      | GW #4's address    | if1    |
   +---------------------+--------------------+--------+
   | 0x0B:0x0D           | GW #3's address    | if0    |
   +---------------------+--------------------+--------+
   | 0x0B:0x0C           | -                  | if1    |
   +---------------------+--------------------+--------+
   | 0x0B                | GW #1's address    | if0    |
   +---------------------+--------------------+--------+
   | 0x0A                | GW #1's address    | if0    |
   +---------------------+--------------------+--------+
                 Figure 8 Example of Routing Table (GW #2)

   +---------------------+--------------------+--------+
   | Destination         | nexthop            | out if |
   +---------------------+--------------------+--------+
   | 0x0B:0x0C           | GW #2's address    | if0    |
   +---------------------+--------------------+--------+
   | 0x0B:0x0D           | -                  | if1    |
   +---------------------+--------------------+--------+
   | 0x0B                | GW #1's address    | if0    |
   +---------------------+--------------------+--------+
   | 0x0A                | GW #1's address    | if0    |
   +---------------------+--------------------+--------+
                 Figure 9 Example of Routing Table (GW #3)

6.2. Server Discovery (path discovery)

   When a requestor gets locator for a specific NDO from NRS, it build a
   path discovery message and forwards to the default domain gateway (in


Lee                    Expires March 17, 2015                [Page 14]

Internet-Draft      Scalable Domain-based Routing       September 2014


   this example, "requestor A" send path discovery message to GW #1).
   The path discovery message includes followings:

   +------------------------------------------------------------------+
   | Network header                                                   |
   +------------------------------------------------------------------+
   | Requestor's ID                                                   |
   +------------------------------------------------------------------+
   | NDO's ID                                                         |
   +------------------------------------------------------------------+
   | NDO's current locator                                            |
   +------------------------------------------------------------------+

   If a domain gateway receives path discovery message, it looks up its
   forwarding cache table firstly. If there is no forwarding cache entry
   for the NDO's ID, the domain gateway builds new forwarding cache
   entry. The domain gateway searches its routing table by using
   "locator (included in the path discovery message)" as key, then it
   builds forwarding cache entry for the NDO's ID and requestor's ID.

   Forwarding cache entry for NDO includes followings:

   o NDO's ID

   o Next hop information (found in the routing table entry)

   Forwarding cache entry for the requestor includes followings:

   o Requestor's ID

   o Previous hop information (found in the network header)

   After building the forwarding cache, the path discovery message is
   forwarded to the next hop domain gateway (network header is changed
   according to the next-hop information of forwarding cache), and
   repeat this procedure until it reaches domain which includes the NDO.

   Forwarding caches are managed in the manner of timer-based way.

6.3. Name-based Forwarding

   When the contents server which has requested NDO receives path
   discovery message, it sends NDO data by using following message
   format:





Lee                    Expires March 17, 2015                [Page 15]

Internet-Draft      Scalable Domain-based Routing       September 2014


   +------------------------------------------------------------------+
   | Network header                                                   |
   +------------------------------------------------------------------+
   | Requestor's ID                                                   |
   +------------------------------------------------------------------+
   | NDO data                                                         |
   +------------------------------------------------------------------+

   Network header would be changed according to the next hop information
   of forwarding cache.

7. Security Considerations

   TBD.

8. IANA Considerations

   TBD.

9. References

9.1. Informative References

   [ICNRG charter] http://irtf.org/icnrg

   [ICN Challenges] D.Kutscher, "ICN Research Channelges", internet-
                     draft, July 2013

   [aRoute] Ahmed, Reaz, Md Faizul Bari, Shihabur Rahman Chowdhury, Md
             Golam Rabbani, Raouf Boutaba, and Bertrand Mathieu.
             "aRoute: A Name Based Routing Scheme for Information
             Centric Networks." In IEEE International Conference on
             Computer Communications (INFOCOM) Mini-Conference, 2013.

   [ICN survey] Ahlgren, Bengt, Christian Dannewitz, Claudio Imbrenda,
               Dirk Kutscher, and Borje Ohlman. "A Survey of
               Information-Centric Networking." Communications Magazine,
               IEEE 50, no. 7 (2012): 26-36.

   [Id net] http://www.idnet.re.kr/

   [BF NRS] J. Hong, "Bloom Filter-based Flat Name Resolution System for
             ICN", http://www.ietf.org/internet-drafts/draft-hong-icnrg-
             bloomfilterbasedname-resolution-00.txt





Lee                    Expires March 17, 2015                [Page 16]

Internet-Draft      Scalable Domain-based Routing       September 2014


   [LSR]   W. Lim, "Design of Scalable Link-State Routing Protocol for
             Efficient Inter-Domain Routing" in IEEE Communication
             letter (is being under examination)

Authors' Addresses

   Joo-Chul Lee
   ETRI
   161 Gajeong-dong, Yuseong-gu, Daejon

   Phone:
   Email: rune@etri.re.kr

   Wan-Seon Lim
   ETRI
   161 Gajeong-dong, Yuseong-gu, Daejon

   Phone:
   Email: wslim@etri.re.kr

   Woo-Jik Chun
   HanKuk University of Foreign Studies
   81, Oedae-ro, Mohyeon-myeon, Cheoin-gu,Yongin-si, Gyeonggi-do, 449-
   791, Korea

   Phone:
   Email: woojikchun@gmail.com





















Lee                    Expires March 17, 2015                [Page 17]