Skip to main content

IPv6 Source Routing for ultralow Latency

The information below is for an old version of the document.
Document Type
This is an older version of an Internet-Draft whose latest revision state is "Active".
Authors Andreas Foglar , Mike Parker , Theodoros Rokkas
Last updated 2018-03-02 (Latest revision 2017-11-12)
RFC stream (None)
Stream Stream state (No stream defined)
Consensus boilerplate Unknown
RFC Editor Note (None)
IESG IESG state I-D Exists
Telechat date (None)
Responsible AD (None)
Send notices to (None)
Routing Area Working Group                         A. Foglar, InnoRoute
INTERNET-DRAFT                                     M. Parker, Uni Essex
Intended status: EXPERIMENTAL                        T. Rokkas, Incites
Expires: September 1, 2018                              March 2, 2018     

          IPv6 Source Routing for ultralow Latency


  This Internet-Draft describes a hierarchical addressing scheme 
  for IPv6, intentionally very much simplified to allow for very 
  fast source routing experimentation using simple forwarding 
  nodes. Research groups evaluate achievable latency reduction 
  for special applications such as radio access networks, 
  industrial networks or other networks requiring very low 
Status of This Memo

  This Internet-Draft is submitted in full conformance with the
  provisions of BCP 78 and BCP 79.
  Internet-Drafts are working documents of the Internet Engineering
  Task Force (IETF).  Note that other groups may also distribute
  working documents as Internet-Drafts.  The list of current Internet-
  Drafts is at
  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."

Copyright Notice

  Copyright (c) 2017 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
  ( 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.
Revision Note for Version 02

  Reference to experimental verification of the concept is added in the
  section "Acknowledgements".  
1. Introduction

  To achieve minimum latency the forwarding nodes must support 
  cut-through technology as opposed to the commonly used store-

  and-forward technology. Cut-through means, that the packet 
  header already leaves a node at the egress port while the tail 
  of the packet is still received at the ingress port. This 
  short time does not allow complex routing decisions. 

  Therefore, a very simple routing address field structure is 
  specified below. It should limit the complexity of the 
  forwarding node used in the experiments. Therefore, in this 
  text the term "forwarding node" is used instead of "router", 
  although the device is operating in OSI Layer 3 and accordingly
  executes router functions such as decrementing the hop limit field.
  Redundancy issues are not considered. 

2. IPv6 address prefix structure

  The following proposal uses the 64-bit IPv6 address prefix.
  Each forwarding node has up to 16 ports and hence needs 4 bits 
  of the address field to decide to which port a packet should 
  be forwarded. The 64-bit prefix is divided into 16 sub-fields 
  of 4 bit, defining up to 16 hierarchy levels. A forwarding 
  node is configured manually to which of the sub-fields it should
  evaluate for the forwarding decision. 
  A number n of leading 4-bit fields cannot be used for forwarding 
  decisions, but must have a special value to indicate the 
  'escape prefix' of the experimental forwarding mode. 
  The 64-bit prefix of the IPv6 address has this structure:
  | n x 4-bit escape prefix |(16-n) x 4-bit address fields |
  The first 4-bit field following the escape prefix has the 
  highest hierarchy level, the last 4-bit field has the lowest 
  hierarchy level.

3. Forwarding node behavior

  The forwarding node has up to 16 downlink ports and at least 
  one uplink port. Typically, the forwarding nodes are arranged 
  in a regular tree structure with one top node, up to 16 nodes 
  in the second hierarchy, up to 256 nodes in the third hierarchy 
  and so on for up to 16-n hierarchies.
  A forwarding node must be configured to operate at a certain 
  position in the hierarchical network. For example, at third 
  hierarchy level, branch 4 of the first hierarchy and branch 12 
  of the second hierarchy. 
  The behavior of each forwarding node is depending on the 
  position of a node in a hierarchical network. For all 
  positions, the first step is to check the escape prefix. Only 
  packets with matching escape prefix are forwarded. 
  The top forwarding node with the highest hierarchy level 
  evaluates the first 4-bit field following the n x 4-bit escape 
  prefix. The value of the evaluation field determines the 
  output port of the packet. The remaining fields are don't 

  | escape prefix | 4-bit | (16-n-1) x 4-bit |
  <  mandatory   > <eval.> <   don't care   > 
  A forwarding node in a lower hierarchy first checks if the 4-
  bit fields preceding the evaluation field match the configured 
  value. In case of match the value of the configured evaluation 
  field of the packet is used as downlink port number where the 
  packet is forwarded. The remaining 4-bit fields are ignored. 
  In case of mismatch the packet is forwarded to the uplink 
  | escape prefix | m x 4-bit | 4-bit | (16-n-m-1) x 4-bit |
  <  mandatory   > <  match  > <eval.> <   don't care     >
  The parameter m indicates the hierarchy level with m=0 
  denoting the highest hierarchy.
  Hence, when a packet enters a hierarchical network at the 
  lowest layer node it is forwarded in uplink direction until it 
  reaches a node where the m x 4-bit prefix matches the 
  configured value of the node. At latest, the highest-level 
  node will always match and forward the packet in the desired 
  downlink direction. 

4. Numerical values

  As mentioned, one pre-requisite of the simple forwarding 
  concept is to keep the complexity of the forwarding nodes low. 
  Also, the configuration of the nodes should be kept simple. In 
  particular industrial networks are operated by persons who are 
  not experts in communication. Configurations should be 
  intuitively understandable by all without long explication. 
  Therefore, for the first experimental forwarding node the 
  number of downlink ports is limited to 10 with numbers 0...9. 16 
  digits at the front panel of the forwarding device show the 
  configuration. Use of classical 7-segment digits make the 
  limits of the configuration obvious. 
  As escape code, the first two digits are fixed to the value 
  "AF" (binary '10101111'). These two characters contrast with 
  the following numerical digits, so that the escape code can be 
  clearly differentiated from the following configuration. The 
  display uses the 'H' character instead of the 'X' the usual 
  term for the variable. 
  The H specifies the digit of the packet prefix which is 
  evaluated for forwarding. When the H is selected all lower 
  digits are automatically set to '-' to indicate the don't care 
  To make the configuration still more obvious it is recommended 
  to configure the local telephone number. With that measure, 
  every local experimentation has unique numbers and can 
  potentially be interconnected via tunnels (IP, MPLS, VPN etc.) 
  with other experimental setups. 
  The length of 14 digits allows sufficient in-house 
  hierarchies, even for industrial applications where forwarding 

  nodes interconnect large numbers of sensor controllers. 
  Inhouse installations would be structured for example in 
  building, floor, fabrication unit, machine - with one sensor 
  controller per machine. For the sake of simplicity numbers are 
  deliberately wasted, for example if the building has only 3 
  stories the digits 4...9 are unused.

5. Example configuration

  A small office in Munich with the telephone number +49-89-
  45241990 configures its local top-level forwarding node to:
  Note that for the sake of simplicity this simplified notation 
  is introduced here as alternative to the usual notation 
  AF49:8945:2419:9000/56. With the new notation, the cabling 
  staff people can immediately check the hierarchy location of 
  the forwarding node and connect the cables to the floors at 
  ports 0...3.
  The next hierarchy level is related to the floor. In case of a 
  3-story building only three next level forwarding nodes are 
  used with these configured values:
      AF49.8945.2419.900H at the ground level
      AF49.8945.2419.901H at the first floor
      AF49.8945.2419.902H at the second floor
      AF49.8945.2419.903H at the third floor.
  In each floor, up to 10 sensor nodes can be connected.
  Each of the sensor nodes can address several sensors/ 
  actuators addressed via the interface identifier contained in 
  the second part of the 128-bit IPv6 address.
  In the following a connection between sensors in this office to 
  otherIoT equipment located in Essex University is described. The
  connection is realized with one additional forwarding node 
  installed at Essex University premises with the second level address
  This high level forwarding node can be used although the phone number
  of the researcher is +44 1206 872413, as long as there is no further 
  node in UK. 
  At downlink port 9 the 13th level forwarding node in Munich is con-
  nected via a  Layer 2  link such as VLAN or SDH pipe or MPLS tunnel.
  The levels in between must not be populated by forwarding nodes as
  long as no other branch is needed at one of the two sides. 
  If for example another site in Munich center must be connected one
  additional forwarding node must be installed with the 5th level

  The small office mentioned above would be connected to downlink port

  4 while the new site would be connected at downlink port 1, the
  prefix for Munich center. The configuration is visualized in the
  Figure below.
    Essex (UK)                     Munich (DE)

  |     AF4H.----.----.---- |
      |                \
      |                 ------ L2 Link ------
  |----------|                               \
  | IoT node |                    |----------U----------|  
  |----------|                    | AF49.89H-.----.---- |
                                     /       \
                                  ---         -----------
                                 /                       \
                     |----------U----------|   |----------U----------|
                     | AF49.89H-.----.---- |   | AF49.8945.2419.90H- |
                     |-0-1-2-3-4-5-6-7-8-9-|   |-0-1-2-3-4-5-6-7-8-9-|
                                        | AF49.8945.5419.901H |   
  U = Uplink                                      |  
                                            | IoT node |
  Figure: Example Configuration 
6. Acknowledgements

  The authors would like to thank the consortium of the European 
  research project CHARISMA for the possibility to experiment. The
  description of the final demonstration is available for download:

7. Authors' Addresses

   Andreas Foglar
   InnoRoute GmbH
   Marsstr. 14a
   80335 Munich

   Mike Parker
   Wivenhoe Park, Colchester
   Essex, CO3 4HG
   United Kingdom

   Theodoros Rokkas
   Incites S.A.R.L.
   130, Route d' Arlon
   Strassen L-8008

Foglar, Parker, Rokkas             Expires September 1, 2018