COINRG                                                          I. Kunze
Internet-Draft                                                 K. Wehrle
Intended status: Informational                    RWTH Aachen University
Expires: 6 May 2021                                           D. Trossen
                                    Huawei Technologies Duesseldorf GmbH
                                                         2 November 2020

                   Use Cases for In-Network Computing


   Computing in the Network (COIN) comes with the prospect of deploying
   functionality on networking devices, such as switches and network
   interface cards.  While such functionality can be beneficial in
   several contexts, it has to be carefully placed into the context of
   the general Internet communication.  This document discusses some use
   cases to demonstrate how real applications can benefit from COIN and
   to showcase essential requirements that have to be fulfilled by COIN

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."

   This Internet-Draft will expire on 6 May 2021.

Copyright Notice

   Copyright (c) 2020 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 (
   license-info) in effect on the date of publication of this document.
   Please review these documents carefully, as they describe your rights

Kunze, et al.              Expires 6 May 2021                   [Page 1]

Internet-Draft               COIN Use Cases                November 2020

   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.

Table of Contents

   1.  Introduction  . . . . . . . . . . . . . . . . . . . . . . . .   2
   2.  Terminology . . . . . . . . . . . . . . . . . . . . . . . . .   3
   3.  Industrial Use Cases  . . . . . . . . . . . . . . . . . . . .   3
     3.1.  IIoT Network Scenario . . . . . . . . . . . . . . . . . .   4
     3.2.  In-Network Control / Time-sensitive applications  . . . .   5
       3.2.1.  Characterization and Requirements . . . . . . . . . .   5
       3.2.2.  Approaches  . . . . . . . . . . . . . . . . . . . . .   6
     3.3.  Large Volume Applications/ Traffic Filtering  . . . . . .   7
       3.3.1.  Characterization and Requirements . . . . . . . . . .   7
       3.3.2.  Approaches  . . . . . . . . . . . . . . . . . . . . .   8
     3.4.  Industrial Safety (Dead Man's Switch) . . . . . . . . . .   9
       3.4.1.  Characterization and Requirements . . . . . . . . . .  10
       3.4.2.  Approaches  . . . . . . . . . . . . . . . . . . . . .  10
   4.  Security Considerations . . . . . . . . . . . . . . . . . . .  11
   5.  Immersive Devices . . . . . . . . . . . . . . . . . . . . . .  11
     5.1.  Mobile Application Offloading . . . . . . . . . . . . . .  11
     5.2.  Edge AR/VR  . . . . . . . . . . . . . . . . . . . . . . .  11
   6.  Infrastructure Services . . . . . . . . . . . . . . . . . . .  11
     6.1.  Distributed AI  . . . . . . . . . . . . . . . . . . . . .  11
     6.2.  Content Delivery Networks . . . . . . . . . . . . . . . .  11
     6.3.  CFaaS . . . . . . . . . . . . . . . . . . . . . . . . . .  12
   7.  Taxonomy  . . . . . . . . . . . . . . . . . . . . . . . . . .  12
   8.  IANA Considerations . . . . . . . . . . . . . . . . . . . . .  12
   9.  Conclusion  . . . . . . . . . . . . . . . . . . . . . . . . .  12
   10. Informative References  . . . . . . . . . . . . . . . . . . .  12
   Authors' Addresses  . . . . . . . . . . . . . . . . . . . . . . .  14

1.  Introduction

   The Internet bases on a best-effort network with limited guarantees
   regarding the timely and successful transmission of packets.
   Functionality is generally provided by the end-hosts while the
   network is kept simple and only intended to forward the packets.
   This design-choice is suitable for general Internet-based
   applications and has helped in the rapid growth of the Internet.
   However, there are several domains which, e.g., demand a number of
   strict performance guarantees that cannot be provided over regular
   best-effort networks.  In this context, flexibly distributing the
   computation tasks across the network can help to achieve the
   guarantees and increase the overall performance.

Kunze, et al.              Expires 6 May 2021                   [Page 2]

Internet-Draft               COIN Use Cases                November 2020

   The different domains, however, have different requirements and it is
   unclear whether there can be a common solution to all COIN scenarios
   or if solutions have to be tailored to each scenario.

   This document first presents applications and requirements of some
   domains to illustrate the importance of COIN for realizing advanced
   applications.  Based on these discussion, the draft then creates a
   taxonomy of COIN scenarios with the goal of guiding future work.

2.  Terminology

   Programmable network devices (PNDs): Network devices, such as network
   interface cards and switches, which are programmable, e.g., using P4

3.  Industrial Use Cases

   The industrial domain is characterized by diverse sets of
   requirements which often cannot be provided over regular best-effort
   networks.  Consequently, there is a large number of specialized
   applications and protocols designed to give the required strict
   performance guarantees, e.g., regarding real-time capabilities.
   Time-Sensitive-Networking [TSN] as an enhancement to the standard
   Ethernet, e.g., tries to achieve these requirements on the link layer
   by statically reserving shares of the bandwidth.  In the Industrial
   Internet of Things (IIoT), however, more and more parts of the
   industrial production domain are interconnected.  This increases the
   complexity of the industrial networks, makes them more dynamic, and
   creates more diverse sets of requirements.  In these scenarios,
   solutions on the link layer alone are not sufficient.

   The challenge is to develop concepts that can satisfy the dynamic
   performance requirements of modern industrial networks.  COIN
   presents a promising starting point because it allows to flexibly
   distribute computation tasks across the network which can help to
   manage dynamic changes.  As specifying general requirements for the
   industrial production domain is difficult due to the mentioned
   diversity, this document next characterizes and analyzes three
   distinct scenarios to showcase potential requirements for the
   industrial production domain, thereby illustrating how COIN can be

Kunze, et al.              Expires 6 May 2021                   [Page 3]

Internet-Draft               COIN Use Cases                November 2020

3.1.  IIoT Network Scenario

   Common components of the IIoT can be divided into three categories as
   illustrated in Figure 1.  Following
   [I-D.mcbride-edge-data-discovery-overview], EDGE DEVICES, such as
   sensors and actuators, constitute the boundary between the physical
   and digital world.  They communicate the current state of the
   physical world to the digital world by transmitting sensor data or
   let the digital world interact with the physical world by executing
   actions after receiving (simple) control information.  The processing
   of the sensor data and the creation of the control information is
   done on COMPUTING DEVICES.  They range from small-powered controllers
   close to the EDGE DEVICES, to more powerful edge or remote clouds in
   larger distances.  The connection between the EDGE and COMPUTING
   DEVICES is established by NETWORKING DEVICES.  In the industrial
   domain, they range from standard devices, e.g., typical Ethernet
   switches, which can interconnect all Ethernet-capable hosts, to
   proprietary equipment with proprietary protocols only supporting
   hosts of specific vendors.

    |Sensor| ------------|              ~~~~~~~~~~~~      ------------
    --------       -------------        { Internet } --- |Remote Cloud|
       .           |Access Point|---    ~~~~~~~~~~~~      ------------
    --------       -------------   |          |
    |Sensor| ----|        |        |          |
    --------     |        |       --------    |
       .         |        |       |Switch| ----------------------
       .         |        |       --------                       |
       .         |        |                   ------------       |
    ----------   |        |----------------- | Controller |      |
    |Actuator| ------------                   ------------       |
    ----------   |    --------                            ------------
       .         |----|Switch|---------------------------| Edge Cloud |
    ----------        --------                            ------------
    |Actuator|  ---------|

   |-----------|       |------------------|     |-------------------|

     Figure 1: Industrial networks show a high level of heterogeneity.

Kunze, et al.              Expires 6 May 2021                   [Page 4]

Internet-Draft               COIN Use Cases                November 2020

3.2.  In-Network Control / Time-sensitive applications

   The control of physical processes and components of a production line
   is essential for the growing automation of production and ideally
   allows for a consistent quality level.  Traditionally, the control
   has been exercised by control software running on programmable logic
   controllers (PLCs) located directly next to the controlled process or
   component.  This approach is best-suited for settings with a simple
   model that is focused on a single or few controlled components.

   Modern production lines and shop floors are characterized by an
   increasing amount of involved devices and sensors, a growing level of
   dependency between the different components, and more complex control
   models.  A centralized control is desirable to manage the large
   amount of available information which often has to be pre-processed
   or aggregated with other information before it can be used.  PLCs are
   not designed for this array of tasks and computations could
   theoretically be moved to more powerful devices.  These devices are
   no longer close to the controlled objects and induce additional

   It is worthwhile to investigate whether the outsourcing of control
   functionality to distant computation platforms is viable because
   these platforms have a high level of flexibility and scalability.  In
   the following, we describe the requirements and characteristics of
   the control setting in more detail.

3.2.1.  Characterization and Requirements

   A control process consists of two main components as illustrated in
   Figure 2: a system under control and a controller.  In feedback
   control, the current state of the system is monitored, e.g., using
   sensors and the controller influences the system based on the
   difference between the current and the reference state to keep it
   close to this reference state.

      state      ------------        --------    Output
   ---------->  | Controller | ---> | System | ---------->
              ^  ------------        --------       |
              |                                     |
              |   observed state                    |
              |                    ---------        |
               -------------------| Sensors | <-----

                  Figure 2: Simple feedback control model.

Kunze, et al.              Expires 6 May 2021                   [Page 5]

Internet-Draft               COIN Use Cases                November 2020

   Apart from the control model, the quality of the control primarily
   depends on the timely reception of the sensor feedback, because the
   controller can only react if it is notified of changes in the system
   state.  Depending on the dynamics of the controlled system, the
   control can be subject to tight latency constraints, often in the
   single-digit millisecond range.  While low latencies are essential,
   there is an even greater need for stable and deterministic levels of
   latency, because controllers can generally cope with different levels
   of latency, if they are designed for them, but they are significantly
   challenged by dynamically changing or unstable latencies.  The
   unpredictable latency of the Internet exemplifies this problem if
   off-premise cloud platforms are included.

   The main requirements for the industrial control scenario are low and
   stable latencies to ensure that processes can work continuously and
   that no machines are damaged.

3.2.2.  Approaches

   Control models, in general, can become involved but there is a
   variety of control algorithms that are composed of simple
   computations such as matrix multiplication.  These are supported by
   some PNDs and it is thus possible to compose simplified
   approximations of the more complex algorithms and deploy them in the
   network.  While the simplified versions induce a more inaccurate
   control, they allow for a quicker response and might be sufficient to
   operate a basic tight control loop while the overall control can
   still be exercised from the cloud.  The problem, however, is that
   networking devices typically only allow for integer precision
   computation while floating-point precision is needed by most control
   algorithms.  Additionally, computational capabilities vary for
   different available PNDs.  Yet, early approaches like [RUETH] and
   [VESTIN] have already shown the general applicability of such ideas,
   but there are still a lot of open research questions not limited to
   the following:

   *  How can one derive the simplified versions of the overall

      -  How complex can they become?

      -  How can one take the limited computational precision of
         networking devices into account when making them?

   *  How does one distribute the simplified versions in the network?

   *  How does the overall controller interact with the simplified

Kunze, et al.              Expires 6 May 2021                   [Page 6]

Internet-Draft               COIN Use Cases                November 2020

3.3.  Large Volume Applications/ Traffic Filtering

   In the IIoT, processes and machines can be monitored more effectively
   resulting in more available information.  This data can be used to
   deploy machine learning (ML) techniques and consequently help to find
   previously unknown correlations between different components of the
   production which in turn helps to improve the overall production
   system.  Newly gained knowledge can be shared between different sites
   of the same company or even between different companies [PENNEKAMP].

   Traditional company infrastructure is neither equipped for the
   management and storage of such large amounts of data nor for the
   computationally expensive training of ML approaches.  Similar to the
   considerations in Section 3.2, off-premise cloud platforms offer
   cost-effective solutions with a high degree of flexibility and
   scalability.  While the unpredictable latency of the Internet is only
   a subordinate problem for this use case, moving all data to off-
   premise locations primarily poses infrastructural challenges which
   are presented in more detail in the following.

3.3.1.  Characterization and Requirements

   Processes in the industrial domain are monitored by distributed
   sensors which range from simple binary (e.g., light barriers) to
   sophisticated sensors measuring the system with varying degrees of
   resolution.  Sensors can further serve different purposes, as some
   might be used for time-critical process control while others are only
   used as redundant fallback platforms.  Overall, there is a high level
   of heterogeneity which makes managing the sensor output a challenging

   Depending on the deployed sensors and the complexity of the observed
   system, the resulting overall data volume can easily be in the range
   of several Gbit/s [GLEBKE].  Using off-premise clouds for managing
   the data requires uploading or streaming the growing volume of sensor
   data using the companies' Internet access which is typically limited
   to a few hundred of Mbit/s.  While large networking companies can
   simply upgrade their infrastructure, most industrial companies rely
   on traditional ISPs for their Internet access.  Higher access speeds
   are hence tied to higher costs and, above all, subject to the supply
   of the ISPs and consequently not always available.  A major challenge
   is thus to devise a methodology that is able to handle such amounts
   of data over limited access links.

   Another aspect is that business data leaving the premise and control
   of the company further comes with security concerns, as sensitive
   information or valuable business secrets might be contained in it.
   Typical security measures such as encrypting the data make in-network

Kunze, et al.              Expires 6 May 2021                   [Page 7]

Internet-Draft               COIN Use Cases                November 2020

   computing techniques hardly applicable as they typically work on
   unencrypted data.  Adding security to in-network computing
   approaches, either by adding functionality for handling encrypted
   data or devising general security measures, is thus an auspicious
   field for research which we describe in more detail in Section 4.

3.3.2.  Approaches

   There are at least two concepts which might be suitable for reducing
   the amount of transmitted data in a meaningful way:

   1.  filtering out redundant or unnecessary data

   2.  aggregating data by applying pre-processing steps within the

   Both concepts require detailed knowledge about the monitoring
   infrastructure at the factories and the purpose of the transmitted
   data.  Traffic Filters

   Sensors are often set up redundantly, i.e., part of the collected
   data might also be redundant.  Moreover, they are often hard to
   configure or not configurable at all which is why their resolution or
   sampling frequency is often larger than required.  Consequently, it
   is likely that more data is transmitted than is needed or desired.  A
   trivial idea for reducing the amount of data is to filter out
   redundant or undesired data before it leaves the premise using simple
   traffic filters that are deployed in the on-premise network.  There
   are different approaches to how this topic can be tackled.  A first
   step would be to scale down the available sensor data to the data
   rate that is needed.  For example, if a sensor transmits with a
   frequency of 5 kHz, but the control entity only needs 1 kHz, only
   every fifth packet containing sensor data is let through.
   Alternatively, sensor data could be filtered down to a lower
   frequency while the sensor value is in an uninteresting range, but
   let through with higher resolution once the sensor value range
   becomes interesting.  It is important that end-hosts are informed
   about the filtering so that they can distinguish between data loss
   and data filtered out on purpose.

   In this context, the following research questions can be of interest:

   *  How can traffic filters be designed?

   *  How can traffic filters be coordinated and deployed?

Kunze, et al.              Expires 6 May 2021                   [Page 8]

Internet-Draft               COIN Use Cases                November 2020

   *  How can traffic filters be changed dynamically?

   *  How can traffic filtering be signaled to the end-hosts?  In-Network (Pre-)Processing

   There are manifold computations that can be performed on the sensor
   data in the cloud.  Some of them are very complex or need the
   complete sensor data during the computation, but there are also
   simpler operations which can be done on subsets of the overall
   dataset or earlier on the communication path as soon as all data is
   available.  One example is finding the maximum of all sensor values
   which can either be done iteratively at each intermediate hop or at
   the first hop, where all data is available.

   Using expert knowledge about the exact computation steps and the
   concrete transmission path of the sensor data, simple computation
   steps can be deployed in the on-premise network to reduce the overall
   data volume and potentially speed up the processing time in the

   Related work has already shown that in-network aggregation can help
   to improve the performance of distributed ML applications [SAPIO].
   Investigating the applicability of stream data processing techniques
   to programmable networking devices is also interesting, because
   sensor data is usually streamed.  In this context, the following
   research questions can be of interest:

   *  Which (pre-)processing steps can be deployed in the network?

      -  How complex can they become?

   *  How can applications incorporate the (pre-)processing steps?

   *  How can the programming of the techniques be streamlined?

3.4.  Industrial Safety (Dead Man's Switch)

   Despite increasing automation in production processes, human workers
   are still often necessary.  Consequently, safety measures have a high
   priority to ensure that no human life is endangered.  In traditional
   factories, the regions of contact between humans and machines are
   well-defined and interactions are simple.  Simple safety measures
   like emergency switches at the working positions are enough to
   provide a decent level of safety.

Kunze, et al.              Expires 6 May 2021                   [Page 9]

Internet-Draft               COIN Use Cases                November 2020

   Modern factories are characterized by increasingly dynamic and
   complex environments with new interaction scenarios between humans
   and robots.  Robots can either directly assist humans or perform
   tasks autonomously.  The intersect between the human working area and
   the robots grows and it is harder for human workers to fully observe
   the complete environment.

   Additional safety measures are essential to prevent accidents and
   support humans in observing the environment.  The increased
   availability of sensor data and the detailed monitoring of the
   factories can help to build additional safety measures if the
   corresponding data is collected early at the correct position.

3.4.1.  Characterization and Requirements

   Industrial safety measures are typically hardware solutions because
   they have to pass rigorous testing before they are certified and
   deployment-ready.  Standard measures include safety switches and
   light barriers.  Additionally, the working area can be explicitly
   divided into 'contact' and 'safe' areas, indicating when workers have
   to watch out for interactions with machinery.

   These measures are static solutions, potentially relying on
   specialized hardware, and are challenged by the increased dynamics of
   modern factories where the factory configuration can be changed on
   demand.  Software solutions offer higher flexibility as they can
   dynamically respect new information gathered by the sensor systems,
   but in most cases they cannot give guaranteed safety.  Yet, it is
   worthwhile to investigate whether such solutions can introduce
   additional safety measures.

3.4.2.  Approaches

   Software-based solutions can take advantage of the large amount of
   available sensor data.  Different safety indicators within the
   production hall can be combined within the network so that
   programmable networking devices can give early responses if a
   potential safety breach is detected.  A rather simple possibility
   could be to track the positions of human workers and robots.
   Whenever a robot gets too close to a human in a non-working area or
   if a human enters a defined safety zone, robots are stopped to
   prevent injuries.  More advanced concepts could also include image
   data or combine arbitrary sensor data.

   In this context, the following research questions can be of interest:

   *  Which additional safety measures can be provided?

Kunze, et al.              Expires 6 May 2021                  [Page 10]

Internet-Draft               COIN Use Cases                November 2020

      -  Do these measures actually improve safety?

   *  Which sensor information can be combined and how?

4.  Security Considerations

   Note: This section will need consolidation once new use cases are
   added to the draft.  Current in-network computing approaches
   typically work on unencrypted plain text data because today's
   networking devices usually do not have crypto capabilities.  As is
   already mentioned in Section 3.3.1, this above all poses problems
   when business data, potentially containing business secrets, is
   streamed into remote computing facilities and consequently leaves the
   control of the company.  Insecure on-premise communication within the
   company and on the shop-floor is also a problem as machines could be
   intruded from the outside.  It is thus crucial to deploy security and
   authentication functionality on on-premise and outgoing communication
   although this might interfere with in-network computing approaches.
   Ways to implement and combine security measures with in-network
   computing are described in more detail in [I-D.fink-coin-sec-priv].

5.  Immersive Devices

5.1.  Mobile Application Offloading

   NOTE: Will be moved here from

5.2.  Edge AR/VR

   NOTE: Could be transfered from (expired)

   Additional potential sources: [I-D.draft-geng-rtgwg-cfn-req-00]

6.  Infrastructure Services

6.1.  Distributed AI

   NOTE: Will be moved here from

6.2.  Content Delivery Networks

   NOTE: Will be moved here from

Kunze, et al.              Expires 6 May 2021                  [Page 11]

Internet-Draft               COIN Use Cases                November 2020

6.3.  CFaaS

   NOTE: Will be moved here from

7.  Taxonomy

   NOTE: The taxonomy is intended to generalize characteristics of the
   different presented use cases and work on it will start once more use
   cases are added to the draft.

8.  IANA Considerations


9.  Conclusion

   There are several domains that can profit from COIN.

   Industrial scenarios have unique sets of requirements mostly focusing
   around tight latency constraints with high required bandwidths.

   NOTE: Further aspects will be added once more use cases are added to
   the draft.

10.  Informative References

   [GLEBKE]   Glebke, R., Henze, M., Wehrle, K., Niemietz, P., Trauth,
              D., Mattfeld MBA, P., and T. Bergs, "A Case for Integrated
              Data Processing in Large-Scale Cyber-Physical Systems",
              Proceedings of the 52nd Hawaii International Conference on
              System Sciences, DOI 10.24251/hicss.2019.871, 2019,

              Geng, L. and P. Willis, "Compute First Networking (CFN)
              Scenarios and Requirements", Work in Progress, Internet-
              Draft, draft-geng-rtgwg-cfn-req-00, 4 November 2019,

              Montpetit, M., "In Network Computing Enablers for Extended
              Reality", Work in Progress, Internet-Draft, draft-
              montpetit-coin-xr-03, 8 July 2019, <

Kunze, et al.              Expires 6 May 2021                  [Page 12]

Internet-Draft               COIN Use Cases                November 2020

              Trossen, D., Sarathchandra, C., and M. Boniface, "In-
              Network Computing for App-Centric Micro-Services", Work in
              Progress, Internet-Draft, draft-sarathchandra-coin-
              appcentres-03, 23 October 2020, <

              Fink, I. and K. Wehrle, "Enhancing Security and Privacy
              with In-Network Computing", Work in Progress, Internet-
              Draft, draft-fink-coin-sec-priv-01, 8 September 2020,

              McBride, M., Kutscher, D., Schooler, E., Bernardos, C.,
              Lopez, D., and X. Foy, "Edge Data Discovery for COIN",
              Work in Progress, Internet-Draft, draft-mcbride-edge-data-
              discovery-overview-05, 1 November 2020,

              Pennekamp, J., Henze, M., Schmidt, S., Niemietz, P., Fey,
              M., Trauth, D., Bergs, T., Brecher, C., and K. Wehrle,
              "Dataflow Challenges in an Internet of Production: A
              Security & Privacy Perspective", Proceedings of the ACM
              Workshop on Cyber-Physical Systems Security & Privacy -
              CPS-SPC'19, DOI 10.1145/3338499.3357357, 2019,

   [RUETH]    Rueth, J., Glebke, R., Wehrle, K., Causevic, V., and S.
              Hirche, "Towards In-Network Industrial Feedback Control",
              Proceedings of the 2018 Morning Workshop on In-Network
              Computing - NetCompute '18, DOI 10.1145/3229591.3229592,
              2018, <>.

   [SAPIO]    Sapio, A., "Scaling Distributed Machine Learning with In-
              Network Aggregation", 2019,

   [TSN]      IEEE, ., "Time-Sensitive Networking (TSN) Task Group",
              2019, <>.

   [VESTIN]   Vestin, J., Kassler, A., and J. Akerberg, "FastReact: In-
              Network Control and Caching for Industrial Control
              Networks using Programmable Data Planes", 2018 IEEE 23rd

Kunze, et al.              Expires 6 May 2021                  [Page 13]

Internet-Draft               COIN Use Cases                November 2020

              International Conference on Emerging Technologies and
              Factory Automation (ETFA), DOI 10.1109/etfa.2018.8502456,
              September 2018,

Authors' Addresses

   Ike Kunze
   RWTH Aachen University
   Ahornstr. 55
   D-52074 Aachen


   Klaus Wehrle
   RWTH Aachen University
   Ahornstr. 55
   D-52074 Aachen


   Dirk Trossen
   Huawei Technologies Duesseldorf GmbH
   Riesstr. 25C
   D-80992 Munich


Kunze, et al.              Expires 6 May 2021                  [Page 14]