RFC 7927






Internet Research Task Force (IRTF)                     D. Kutscher, Ed.
Request for Comments: 7927                                           NEC
Category: Informational                                           S. Eum
ISSN: 2070-1721                                         Osaka University
                                                          K. Pentikousis
                                                              Travelping
                                                               I. Psaras
                                                                     UCL
                                                               D. Corujo
                                                  Universidade de Aveiro
                                                               D. Saucez
                                                                   INRIA
                                                              T. Schmidt
                                                             HAW Hamburg
                                                            M. Waehlisch
                                                               FU Berlin
                                                               July 2016


        Information-Centric Networking (ICN) Research Challenges

Abstract



   This memo describes research challenges for Information-Centric
   Networking (ICN), an approach to evolve the Internet infrastructure
   to directly support information distribution by introducing uniquely
   named data as a core Internet principle.  Data becomes independent
   from location, application, storage, and means of transportation,
   enabling or enhancing a number of desirable features, such as
   security, user mobility, multicast, and in-network caching.
   Mechanisms for realizing these benefits is the subject of ongoing
   research in the IRTF and elsewhere.  This document describes current
   research challenges in ICN, including naming, security, routing,
   system scalability, mobility management, wireless networking,
   transport services, in-network caching, and network management.

   This document is a product of the IRTF Information-Centric Networking
   Research Group (ICNRG).













Kutscher, et al.              Informational                     [Page 1]

RFC 7927                     ICN Challenges                    July 2016


Status of This Memo



   This document is not an Internet Standards Track specification; it is
   published for informational purposes.

   This document is a product of the Internet Research Task Force
   (IRTF).  The IRTF publishes the results of Internet-related research
   and development activities.  These results might not be suitable for
   deployment.  This RFC represents the consensus of the Information-
   Centric Networking Research Group of the Internet Research Task Force
   (IRTF).  Documents approved for publication by the IRSG are not a
   candidate for any level of Internet Standard; see Section 2 of RFC
   7841.

   Information about the current status of this document, any errata,
   and how to provide feedback on it may be obtained at
   http://www.rfc-editor.org/info/rfc7927.

Copyright Notice



   Copyright (c) 2016 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.






















Kutscher, et al.              Informational                     [Page 2]

RFC 7927                     ICN Challenges                    July 2016


Table of Contents



   1. Introduction ....................................................4
   2. Problems with Host-Centric Communications .......................4
   3. ICN Terminology and Concepts ....................................6
      3.1. Terminology ................................................6
      3.2. Concepts ...................................................6
   4. ICN Research Challenges .........................................8
      4.1. Naming, Data Integrity, and Data Origin Authentication .....8
      4.2. Security ..................................................10
           4.2.1. Data Integrity and Origin Authentication ...........10
           4.2.2. Binding NDOs to Real-World Identities ..............11
           4.2.3. Access Control and Authorization ...................12
           4.2.4. Encryption .........................................13
           4.2.5. Traffic Aggregation and Filtering ..................13
           4.2.6. State Overloading ..................................13
           4.2.7. Delivering Data Objects from Replicas ..............14
           4.2.8. Cryptographic Robustness ...........................14
           4.2.9. Routing and Forwarding Information Bases ...........15
      4.3. Routing and Resolution System Scalability .................15
           4.3.1. Route-By-Name Routing ..............................15
           4.3.2. Lookup-By-Name Routing .............................16
           4.3.3. Hybrid Routing .....................................17
      4.4. Mobility Management .......................................18
      4.5. Wireless Networking .......................................20
      4.6. Rate and Congestion Control ...............................22
      4.7. In-Network Caching ........................................24
           4.7.1. Cache Placement ....................................24
           4.7.2. Content Placement: Content-to-Cache Distribution ...25
           4.7.3. Request-to-Cache Routing ...........................26
           4.7.4. Staleness Detection of Cached NDOs .................26
           4.7.5. Cache Sharing by Multiple Applications .............27
      4.8. Network Management ........................................27
      4.9. ICN Applications ..........................................29
           4.9.1. Web Applications ...................................30
           4.9.2. Video Streaming and Download .......................30
           4.9.3. Internet of Things .................................31
   5. Security Considerations ........................................32
   6. Informative References .........................................32
   Acknowledgments ...................................................37
   Authors' Addresses ................................................37










Kutscher, et al.              Informational                     [Page 3]

RFC 7927                     ICN Challenges                    July 2016


1.  Introduction



   Information-Centric Networking (ICN) is an approach to evolve the
   Internet infrastructure to directly support accessing Named Data
   Objects (NDOs) as a first-order network service.  Data objects become
   independent of location, application, storage, and means of
   transportation, allowing for inexpensive and ubiquitous in-network
   caching and replication.  The expected benefits are improved
   efficiency and security, better scalability with respect to
   information/bandwidth demand, and better robustness in challenging
   communication scenarios.

   ICN concepts can be deployed by retooling the protocol stack: name-
   based data access can be implemented on top of the existing IP
   infrastructure, e.g., by allowing for named data structures,
   ubiquitous caching, and corresponding transport services, or it can
   be seen as a packet-level internetworking technology that would cause
   fundamental changes to Internet routing and forwarding.  In summary,
   ICN can evolve the Internet architecture towards a network model
   based on named data with different properties and different services.

   This document presents the ICN research challenges that need to be
   addressed in order to achieve these goals.  These research challenges
   are seen from a technical perspective, although business
   relationships between Internet players will also influence
   developments in this area.  We leave business challenges for a
   separate document, however.  The objective of this memo is to
   document the technical challenges and corresponding current
   approaches and to expose requirements that should be addressed by
   future research work.

   This document has been reviewed, commented on, and discussed
   extensively for nearly two years by the vast majority of ICNRG
   members, which certainly exceeds 100 individuals.  It is the
   consensus of ICNRG that the research challenges described in this
   document should be published in the IRTF stream of the RFC series.
   This document does not constitute a standard.

2.  Problems with Host-Centric Communications



   The best current practice to manage the above-mentioned growth in
   terms of data volume and number of devices is to increase
   infrastructure investment, employ application-layer overlays that
   cache content such as Content Distribution Networks (CDNs) and Peer-
   to-Peer (P2P) applications, provide location-independent access to
   data, and optimize its delivery.  In principle, such platforms





Kutscher, et al.              Informational                     [Page 4]

RFC 7927                     ICN Challenges                    July 2016


   provide a service model of accessing named data objects (NDOs) (e.g.,
   replicated web resources in data centers) instead of a host-to-host
   packet delivery service model.

   However, since this functionality resides in overlays only, the full
   potential of content distribution platforms cannot be leveraged as
   the network is not aware of data requests and data transmissions.
   This has the following impact:

   o  Data traffic typically follows sub-optimal paths as it is
      effectively routed, depending on the overlay topology instead of
      the Internet-layer topology.

   o  Network capabilities, such as multicast and broadcast, are largely
      underutilized or not employed at all.  As a result, request and
      delivery for the same object have to be made multiple times.

   o  Overlays typically require significant infrastructure support,
      e.g., authentication portals, content storage, and applications
      servers, making it often impossible to establish local, direct
      communication.

   o  The forwarding layer cannot cooperate with transport-layer
      functions, so sometimes useful functionality such as local
      retransmission and local rate control have to be implemented with
      TCP proxies or other intermediaries.

   o  Provenance validation uses host authentication today.  As such,
      even if there are locally cached copies available, it is normally
      not easily possible to validate their authenticity.

   o  Many applications follow their own approach to caching,
      replication, transport, and authenticity validation (if at all),
      although they all share similar models for accessing named data
      objects in the network.

   Host-centric communication systems restrict applications to data
   transfer between end-hosts only.  Naming data directly provides a
   powerful "hook" for applications to exploit and natively support
   multi-party communication, e.g., multi-source/multi-destination
   communication and a ubiquitous information ecosystem that is not
   restricted to end-host addresses.









Kutscher, et al.              Informational                     [Page 5]

RFC 7927                     ICN Challenges                    July 2016


3.  ICN Terminology and Concepts



3.1.  Terminology



   Information-Centric Networking (ICN):  A concept for communicating in
      a network that provides accessing named data objects as a first
      order service.  See Section 3.2 for details.

   Named Data Object (NDO):  Addressable data unit in an information-
      centric network that can represent a collection of bytes or a
      piece of information.  In ICN, each data object has a name bound
      to it, and there are typically mechanisms to secure (and validate)
      this binding.  Different ICN approaches have different concepts
      for how to map NDOs to individual units of transport, e.g., chunks
      and segments.  Sometimes smaller units may be represented by NDOs
      themselves.  Within the context of this document, an NDO is any
      named data object that can be requested from the network, and we
      do not consider sub-units below the NDO level.  In this document,
      we often use the terms "NDO" and "data object" interchangeably.

   Requestor:  Entity in an ICN network that is sending a request for a
      named data object to the network.

   Publisher:  Entity in an ICN network that publishes an NDO to the
      network, so that corresponding requests can reach the publisher.
      The publisher does not need to be identical to the actual creator,
      for example, a publisher could provide the service of hosting NDOs
      on behalf of the actual creators/owners.

3.2.  Concepts



   Fundamentally, ICN provides access to named data as a first-order
   network service, i.e., the network is able to serve requests to named
   data natively.  That means network nodes can receive requests for
   named data and act as necessary, for example, by forwarding the
   request to a suitable next hop.  Consequently, the network processes
   requests for named data objects (and corresponding responses)
   natively.  Every network node on a path is enabled to perform
   forwarding decisions, cache objects, etc.  This enables the network
   to forward such requests on optimal paths, employing the best
   transmission technologies at every node, e.g., broadcast/multicast
   transmission in wireless networks to avoid duplicate transmission of
   both requests and responses.

   In ICN, there is a set of common concepts and node requirements
   beyond this basic service model.  Naming data objects is a key
   concept.  In general, ICN names represent neither network nodes nor
   interfaces -- they represent NDOs independently of their location.



Kutscher, et al.              Informational                     [Page 6]

RFC 7927                     ICN Challenges                    July 2016


   Names do play a key role in forwarding decisions and are used for
   matching requests to responses: in order to provide better support
   for accessing copies of NDOs regardless of their location, it is
   important to be able to validate that a response actually delivers
   the bits that correspond to an original request for named data.

   Name-content binding validation is a fundamental security service in
   ICN, and this is often achieved by establishing a verifiable binding
   between the object name and the actual object or an identity that has
   created the object.  ICN can support other security services, such as
   provenance validation and encryption, depending on the details of
   naming schemes, object models, and assumptions on infrastructure
   support.  Security services such as name-content binding validation
   are available to any node, i.e., not just the actual requestors.
   This is an important feature for enabling ingress gateways to check
   object authenticity to prevent denial-of-service attacks.

   Based on these fundamental properties, it is possible to leverage
   network storage ubiquitously: every ICN node can cache data objects
   and respond to requests for such objects -- it is not required to
   validate the authenticity of the node itself since name-content
   bindings can be validated.  Ubiquitous in-network storage can be used
   for different purposes: it can enable sharing, i.e., the same object
   copy can be delivered to multiple users/nodes as in today's proxy
   caches and CDNs.  It can also be used to make communication more
   robust (and perform better) by enabling the network to answer
   requests from local caches (instead of from origin servers).  In case
   of disruption (message not delivered), a node can resend the request,
   and it could be answered by an on-path cache, i.e., on the other side
   of the disrupted link.  The network itself would be able to send
   local retransmissions, which enables shorter round-trip times and the
   offloading of origin servers and other parts of the network.

   ICN potentially retrieves segments of NDOs from multiple data
   sources, so only a requestor can determine the completion of a
   retrieval process, i.e., the retrieval of NDOs or individual segments
   is typically controlled by a requestor.  For this reason, ICN
   transport protocols are typically based on a receiver-driven
   mechanism: requestors can control message sending rates by regulating
   the request sending rate (assuming that every response message has to
   be triggered by a request message).  Retransmission would be achieved
   by resending requests, e.g., after a timeout.  Because objects can be
   replicated, object transmission and transport sessions would not
   necessarily have end-to-end semantics: requests can be answered by
   caches, and a node can select one or multiple next-hop destinations
   for a particular request depending on configuration, observed
   performance, or other criteria.




Kutscher, et al.              Informational                     [Page 7]

RFC 7927                     ICN Challenges                    July 2016


   This receiver-driven communication model potentially enables new
   interconnection and business models: a request for named data can be
   linked to an interest of a requestor (or requesting network) in data
   from another peer, which could suggest modeling peering agreements
   and charging accordingly.

4.  ICN Research Challenges



4.1.  Naming, Data Integrity, and Data Origin Authentication



   Naming data objects is as important for ICN as naming hosts is for
   today's Internet.  Fundamentally, ICN requires unique names for
   individual NDOs, since names are used for identifying objects
   independently of their location or container.  In addition, since
   NDOs can be cached anywhere, the origin cannot be trusted anymore,
   hence the importance of establishing a verifiable binding between the
   object and its name (name-data binding validation) so that a
   requestor can be sure that the received bits do correspond to the NDO
   originally requested (data integrity).  Data origin authentication is
   a different security service that can be related to naming, i.e.,
   verifying that an NDO has indeed been published by a publisher (that
   could be identified by a name prefix).

   The above functions are fundamentally required for the information-
   centric network to work reliably; otherwise, neither network elements
   nor requestors can trust object authenticity.  Lack of this trust
   enables several attacks, including DoS attacks, by injecting spoofed
   content into the network.  There are different ways to use names and
   cryptography to achieve the desired functions [ICNNAMING]
   [ICNSURVEY], and there are different ways to manage namespaces
   correspondingly.

   Two types of naming schemes have been proposed in the ICN literature:
   hierarchical and flat namespaces.  For example, a hierarchical scheme
   may have a structure similar to current URIs, where the hierarchy is
   rooted in a publisher prefix.  Such hierarchy enables aggregation of
   routing information, improving scalability of the routing system.  In
   some cases, names are human readable, which makes it possible for
   users to manually type in names, reuse, and, to some extent, map the
   name to the user intent.

   The second general class of naming schemes enables verifying the
   object's name-data integrity without requiring a Public Key
   Infrastructure (PKI) or other third party to first establish trust in
   the key.  This is achieved, e.g., by binding the hash of the NDO
   content to the object's name.  For instance, this can be done by
   directly embedding the hash of the content in the name.  Another
   option is an indirect binding, which embeds the public key of the



Kutscher, et al.              Informational                     [Page 8]

RFC 7927                     ICN Challenges                    July 2016


   publisher in the name and signs the hash of the content with the
   corresponding private key.  The resulting names are typically non-
   hierarchical, or flat, although the publisher field could be employed
   to create a structure that could facilitate route aggregation.

   There are several design trade-offs for ICN naming that affect
   routing and security.  Hash-based names are not human readable nor
   hierarchical.  They can, however, provide some structure for
   aggregation, for instance, a name part corresponding to a publisher.
   In hash-based names with indirect binding, the key of the publisher
   is bound to the name of NDO, so when a user receives, e.g., the
   triplet, namely (data, key, signature), the receiving entity can
   verify that the NDO has been generated by the possessor of the
   private/public key pair and that the NDO has not been changed in
   transit (data integrity).  This can be done by cryptographically
   hashing the received key and the name of the NDO, and comparing it
   with the received hashed key.  Then, the key can be used to verify
   the signature.

   Data origin authentication can be achieved by validating signatures
   based on public key cryptography about an NDO's name and content.  In
   order to ascertain data integrity and origin authenticity with such
   an approach, a PKI-like system is required that would allow linking
   the corresponding public key to a trust chain.

   Research challenges specific to naming include:

   o  Naming static data objects can be performed by using content
      hashes as part of object names, so that publishers can calculate
      the hash over existing data objects and requestors, and any ICN
      node can validate the name-content binding by recalculating the
      hash and comparing it to the name (component).  [RFC6920]
      specifies a concrete naming format for this.

   o  Naming dynamic objects refers to use cases where the name has to
      be generated before the object is created.  For example, this
      could be the case for live streaming, when a publisher wants to
      make the stream available by registering stream chunk names in the
      network.  One approach to this can be hash-based names with
      indirect binding as described above.

   o  Requestor privacy protection can be a challenge in ICN as a direct
      consequence of the accessing-named-data-objects paradigm: if the
      network can "see" requests and responses, it can also log request
      history for network segments or individual users, which can be
      undesirable, especially since names are typically expected to be





Kutscher, et al.              Informational                     [Page 9]

RFC 7927                     ICN Challenges                    July 2016


      long-lived.  That is, even if the name itself does not reveal much
      information, the assumption is that the name can be used to
      retrieve the corresponding data objects in the future.

   o  Updating and versioning NDOs can be challenging because it can
      contradict fundamental ICN assumptions: if an NDO can be
      replicated and stored in in-network storage for later retrieval,
      names have to be long-lived and the name-content binding must not
      change; updating an object (i.e., changing the content without
      generating a new name) is not possible.  Versioning is one
      possible solution but requires a naming scheme that supports it
      (and a way for requestors to learn about newer and older
      versions).

   o  Managing accessibility can also be a challenge.  In ICN, the
      general assumption is to enable ubiquitous access to NDOs, but
      there can be relevant use cases where access to objects should be
      restricted, for example, to a specific user group.  There are
      different approaches for this, such as object encryption
      (requiring key distribution and related mechanisms) or the concept
      of scopes, e.g., based on names that can only be used/resolved
      under some constraints.

4.2.  Security



   Security is an active research field in ICN.  This section provides
   an overview of important security features and corresponding
   challenges that are related to shifting to information-centric
   communications.  Some challenges are well understood, and there are
   (sometimes multiple different) approaches to address them, whereas
   other challenges are active research and engineering topics.

4.2.1.  Data Integrity and Origin Authentication



   As mentioned in Section 4.1, data integrity verification is an
   important ICN feature, since NDOs are retrieved not only from an
   original copy holder but also from any caching point.  Hence, the
   communication channel endpoints to retrieve NDOs are not trustable
   anymore, and solutions widely used today such as Transport Layer
   Security (TLS) [RFC5246] cannot be used as a general solution.  Since
   data objects can be maliciously modified, ICN should provide
   receivers with a security mechanism to verify the integrity of the
   data object, and there are different ways to achieve this.








Kutscher, et al.              Informational                    [Page 10]

RFC 7927                     ICN Challenges                    July 2016


   An efficient approach for static NDOs is providing a name-content-
   binding by hashing an NDO and using the hash as a part of the
   object's name.  [RFC6920] provides a mechanism and a format for
   representing a digest algorithm and the actual digest in a name
   (amongst other information).

   For dynamic objects where it is desirable to refer to an NDO by name
   before the object has been created, public key cryptography is often
   applied, i.e., every NDO would be authenticated by means of a
   signature performed by the data object publisher so that any data
   object consumer can verify the validity of the data object based on
   the signature.  However, in order to verify the signature of an
   object, the consumer must know the public key of the entity that
   signed the object.

   Data origin authentication, i.e., verifying that an NDO has indeed
   been published by a publisher, requires a secure binding of an NDO
   name to a publisher identity -- this is also typically implemented
   using public key cryptography, i.e., by requiring a receiver to
   verify digital signatures that are part of received messages.

   One research challenge is then to support a mechanism to distribute
   the publisher's public keys to the consumers of data objects.  There
   are two main approaches to achieve this: one is based on an external
   third-party authority such as hierarchical Public Key Infrastructure
   (PKI) (see [RFC5280] for a description of hierarchical PKI), and the
   other is to adapt a hash-based scheme with indirect binding.  The
   former, as the name implies, depends on an external third party
   authority to distribute the public key of the publisher for the
   consumers.  In a hash-based scheme with indirect binding, the public
   key (or a hash of it) would be used as part of the name -- which is
   sufficient to validate the data integrity.

   In cases where information about the origin of a data object is not
   available by other means, the object itself would have to incorporate
   the necessary information to determine the object publisher, for
   example, with a certificate, that can be validated through the PKI.
   Once the certificate is authenticated, its public key can be used to
   authenticate the signed data object itself.

4.2.2.  Binding NDOs to Real-World Identities



   In addition to validating NDO authenticity, it is still important to
   bind real-world identities, e.g., a publisher identity, to objects,
   so that a requestor can verify that a received object was actually
   published by a certain source.





Kutscher, et al.              Informational                    [Page 11]

RFC 7927                     ICN Challenges                    July 2016


   With hash-based names, real-world identity bindings are not
   intrinsically established: the name provides the hash of the NDO or
   of the public key that was used to sign the NDO.  There needs to be
   another binding to a real-world identity if that feature is
   requested.

   If the object name directly provides the publisher name and if that
   name is protected by a certificate that links to a PKI-like trust
   chain, the object name itself can provide an intrinsic binding to a
   real-world identity.

   Binding between NDOs and real-world identities is essential, but
   there is no universal way to achieve it as it is all intrinsic to a
   particular ICN approach.

4.2.3.  Access Control and Authorization



   Access control and authorization is a challenge in ICN, because of
   the lack of user-to-server authentication in the fundamental
   communication model based on named data.

   All ICN entities are capable of delivering NDOs on demand due to
   their in-network caching function.  In such an environment,
   traditional access control schemes based on Access Control List (ACL)
   are ill-suited since widely distributed ICN entities have to maintain
   an identical control policy over NDOs for each consumer, which is
   prohibited due to computational overhead and privacy issues.  There
   are two complementary approaches to address the issues:

   1.  Separated approach: access control service from a third party
       that is independent from ICN entities.  Due to the clear
       separation, ICN entities are free from computational overhead to
       determine the accessibility of NDOs by consumers; also, consumers
       can secure their privacy through the independent authorization
       entity [ACCESS-CTL-DEL].  Relevant challenges to this approach
       include reducing the authorization delay (when communicating to
       the access control provider) and currency and consistency of
       access control information (when access control lists are
       distributed).

   2.  Integrated approach: access control service from ICN entities.
       This mechanism is often based on content encryption and key
       distribution [ENCRYPTION-AC].  As mentioned previously, this
       approach suffers from prohibitive overhead for ICN entities due
       to the process of key verification.  While key distribution is
       challenging per se, this approach is beneficial in a way that
       NDOs can be retrieved without the help of an external access
       control provider.  Challenges to this approach include:



Kutscher, et al.              Informational                    [Page 12]

RFC 7927                     ICN Challenges                    July 2016


       1.  applying an access control mechanism for dynamic NDOs in in-
           network caches in a timely manner;

       2.  providing consumers with the different levels of
           accessibility to individual NDOs in a scalable manner; and

       3.  managing key revocation and similar PKI management functions.

4.2.4.  Encryption



   In ICN, NDOs can be encrypted to implement access control (only
   consumers in possession of corresponding decryption keys can access
   the content) or privacy (same approach).  Distributing and managing
   the corresponding keys as well as providing usable interfaces to
   applications and human users are challenges and the subject of
   ongoing work.

   In principle, the challenges are similar to those of broadcast/media
   distribution, and similar approaches (combing symmetric with public
   key cryptography) are being investigated [NDN-CTL-SHARING].

4.2.5.  Traffic Aggregation and Filtering



   One request message to retrieve a data object can actually aggregate
   requests coming from several consumers.  This aggregation of requests
   reduces the overall traffic but makes per-requestor filtering harder.
   The challenge in this case is to provide a mechanism that allows
   request aggregation and per-requestor filtering.  A possible solution
   is to indicate the set of requestors in the aggregated request such
   that the response can indicate the subset of requestors allowed to
   access the data object.  However, this solution requires
   collaboration from other nodes in the network and is not suitable for
   caching.  Another possible solution is to encrypt data objects and
   ensure that only authorized consumers can decrypt them.  This
   solution does not preclude caching and does not require collaboration
   from the network.  However, it implies a mechanism to generate group
   keys (e.g., different private keys can be used to decrypt the same
   encrypted data object) [CHAUM].

4.2.6.  State Overloading



   ICN solutions that implement state on intermediate routers for
   request routing or forwarding (e.g., Content-Centric Networking (CCN)
   [CCN]) are subject to denial-of-service attacks from overloading or
   superseding the internal state of a router (e.g., "interest flooding"
   [BACKSCATTER]).  Additionally, stateful forwarding can enable attack
   vectors such as resource exhaustion or complexity attacks to the
   routing infrastructure.  The challenge is then to provision routers



Kutscher, et al.              Informational                    [Page 13]

RFC 7927                     ICN Challenges                    July 2016


   and construct internal state in a way that alleviates sensibility to
   such attacks.  The problem becomes even harder if the protocol does
   not provide information about the origin of messages.  Without
   origin, it is a particular challenge to distinguish between regular
   (intense) use and misuse of the infrastructure.

4.2.7.  Delivering Data Objects from Replicas



   A common capability of ICN solutions is data replication and in-
   network storage.  Delivering replicated data objects from caches
   decouples content consumption from data sources, which leads to a
   loss of control on (1) content access and (2) content dissemination.
   In a widely distributed, decentralized environment like the Internet,
   this raises several challenges.

   One group of challenges is related to content management.  Without
   access control, a content provider loses the means to count and
   survey content consumption, to limit access scopes, to control or
   know about the number of copies of its data in the network, or to
   withdraw earlier publications reliably.  Any non-cooperative or
   desynchronized data cache may hinder an effective content management
   policy.

   Another group of challenges arises from potential traffic
   amplifications in the decoupled environment.  ICN solutions that
   attempt to retrieve content from several replicas in parallel, or
   decorrelated network routing states, but also distributed attackers
   may simultaneously initiate the transmission of content from multiple
   replicas towards the same destination (e.g., "initiated overloads" or
   "blockades" [BACKSCATTER]).  Methods for mitigating such threats need
   rigorous forwarding checks that require alignment with caching
   procedures (e.g., on-path or off-path).

4.2.8.  Cryptographic Robustness



   Content producers sign their content to ensure the integrity of data
   and to allow for data object authentication.  This is a fundamental
   requirement in ICN due to distributed caching.  Publishers, who
   massively sign content, which is long-lived, offer time and data to
   an attacker for comprising cryptographic credentials.  Signing a
   large amount of data eases common attacks that try to breach the key
   of the publisher.  Based on this observation, the following research
   challenges emerge:








Kutscher, et al.              Informational                    [Page 14]

RFC 7927                     ICN Challenges                    July 2016


   o  To which extent does the content publication model conflict with
      the cryptographic limitations?

   o  How can we achieve transparent re-signing without introducing
      additional cryptographic weaknesses or key management overhead?

   In general, ICN implementations should be designed considering the
   guidelines provided by [RFC7696], especially regarding cryptographic
   algorithm agility, for example, [RFC6920] specifies a naming scheme
   for hash-based names that was designed to support algorithm agility.

4.2.9.  Routing and Forwarding Information Bases



   In information-centric networks, one attack vector is to increase the
   size of routing and forwarding information bases at ICN nodes, i.e.,
   attacking routing scalability in networks that rely on routing by
   name.  This is an intrinsic ICN security issue: possible mitigation
   approaches include combining routing information authenticity
   validation with filtering (e.g., maximum de-aggregation level
   whenever applicable, blacklists, etc.,).

4.3.  Routing and Resolution System Scalability



   ICN routing is a process that finds an NDO based on its name
   initially provided by a requestor.  ICN routing may comprise three
   steps: (1) name resolution, (2) discovery, and (3) delivery.  The
   name resolution step translates the name of the requested NDO into
   its locator.  The discovery step routes the request to the data
   object based on its name or locator.  The last step (delivery) routes
   the data object back to the requestor.  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) as discussed in the following subsections.

4.3.1.  Route-By-Name Routing



   RBNR omits the first name resolution step as the name of the NDO is
   directly used to route the request to the data object.  Therefore,
   routing information for each data object has to be maintained in the
   routing table.  Since the number of data objects is very large
   (estimated as 10^11 back in 2007 [DONA], but this may be
   significantly larger than that, e.g., 10^15 to 10^22), the size of
   routing tables becomes a concern, as it can be proportional to the
   number of data objects unless an aggregation mechanism is introduced.
   On the other hand, RBNR reduces overall latency and simplifies the
   routing process due to the omission of the resolution process.  For
   the delivery step, RBNR needs another identifier (ID) of either host
   or location to forward the requested data object back to the



Kutscher, et al.              Informational                    [Page 15]

RFC 7927                     ICN Challenges                    July 2016


   requestor.  Otherwise, an additional routing mechanism has to be
   introduced, such as breadcrumbs routing [BREADCRUMBS], in which each
   request leaves behind a trail of breadcrumbs along its forwarding
   path, and then the response is forwarded back to the requestor
   consuming the trail.

   Challenges specific to RBNR include:

   o  How can we aggregate the names of data objects to reduce the
      number of routing entries?

   o  How does a user learn the name that is designed for aggregation by
      the provider?  For example, although we name our contribution as
      "ICN research challenges", the IRTF (provider) may want to change
      the name to "/IETF/IRTF/ICN/Research challenges" for aggregation.
      In this case, how does a user learn the name "/IETF/IRTF/ICN/
      Research challenges" to retrieve the contribution initially named
      "ICN research challenges" without any resolution process?

   o  Without introducing the name aggregation scheme, can we still
      achieve scalable routing by taking advantage of topological
      structure and distributed copies?  For example, would employing
      compact routing [COMPACT], random walk [RANDOM], or greedy routing
      [GREEDY] work at the Internet scale?

   o  How can we incorporate copies of a data object in in-network
      caches in this routing scheme?

   o  Breadcrumbs routing implies a symmetric path for ICN request and
      response delivery.  Some network configurations and link types
      prohibit symmetric path forwarding, so it would be challenging to
      interconnect such networks to an infrastructure based on
      breadcrumbs routing.  For example, certain forwarding strategies
      in Delay-Tolerant Networking (DTN) [RFC4838] are employing
      opportunistic forwarding where responses cannot be assumed to
      travel the same path as requests.

4.3.2.  Lookup-By-Name Routing



   LBNR uses the first name resolution step to translate the name of the
   requesting data object into its locator.  Then, the second discovery
   step is carried out based on the locator.  Since IP addresses could
   be used as locators, the discovery step can depend on the current IP
   infrastructure.  The delivery step can be implemented similarly to IP
   routing.  The locator of the requestor is included in the request
   message, and then the requested data object is delivered to the
   requestor based on the locator.  An instantiation of LBNR is [MDHT].




Kutscher, et al.              Informational                    [Page 16]

RFC 7927                     ICN Challenges                    July 2016


   Challenges specific to LBNR include:

   o  How can we build a scalable resolution system that provides:

      *  Fast lookup: Mapping the name of a data object to its locators
         (copies as well).

      *  Fast update: The location of a data object is expected to
         change frequently.  Also, multiple data objects may change
         their locations at the same time, e.g., data objects in a
         laptop.

   o  How can we incorporate copies of a data object in in-network
      caches in this routing scheme?

4.3.3.  Hybrid Routing



   HR combines RBNR and LBNR to benefit from their advantages.  Within a
   single administrative domain, e.g., an ISP, where scalability issues
   can be addressed with network planning, RBNR can be adopted to reduce
   overall latency by omitting the resolution process.  On the other
   hand, LBNR can be used to route between domains that have their own
   prefix (locator).

   For instance, a request message initially includes the name of the
   NDO for the operation of RBNR and is forwarded to a cached copy of
   the NDO or the original server.  When the request message fails to
   find a routing entry in the router, a name resolution step kicks in
   to translate the name into its locator before forwarding the request
   message based on the retrieved locator.

   Challenges specific to HR are:

   o  How can we design a scalable mapping system that, given the name
      of the NDO, should return a destination domain locator so that a
      user request can be encapsulated and forwarded to the domain?

   o  How can the mapping information be secured to prevent a malicious
      router from hijacking the request message by chaining its locator?

   o  How can the bind between the name and the content of the NDO be
      maintained for the verification of its origin and integrity when
      the name changes due to the retrieved locator?








Kutscher, et al.              Informational                    [Page 17]

RFC 7927                     ICN Challenges                    July 2016


4.4.  Mobility Management



   Mobility management has been an active field in host-centric
   communications for more than two decades.  In IETF in particular,
   starting with [RFC2002], a multitude of enhancements to IP have been
   standardized aiming to "allow transparent routing of IP datagrams to
   mobile nodes in the Internet" [RFC5944].  In a nutshell, mobility
   management for IP networks is locator-oriented and relies on the
   concept of a mobility anchor as a foundation for providing always-on
   connectivity to mobile nodes (see [MMIN]).  Other standards
   organizations, such as 3GPP, have followed similar anchor-based
   approaches.  Traffic to and from the mobile node must flow through
   the mobility anchor, typically using a set of tunnels, enabling the
   mobile node to remain reachable while changing its point of
   attachment to the network.

   Needless to say, an IP network that supports node mobility is more
   complex than one that does not, as specialized network entities must
   be introduced in the network architecture.  This is reflected in the
   control plane as well, which carries mobility-related signaling
   messages, establishes and tears down tunnels, and so on.  While
   mobile connectivity was an afterthought in IP, in ICN, this is
   considered a primary deployment environment.  Most, if not all, ICN
   proposals consider mobility from the very beginning, although at
   varying levels of architectural and protocol detail.  That said, no
   solution has so far come forward with a definite answer on how to
   handle mobility in ICN using native primitives.  In fact, we observe
   that mobility appears to be addressed on an ICN proposal-specific
   basis.  That is, there is no single paradigm solution, akin to
   tunneling through a mobility anchor in host-centric networking, that
   can be applied across different ICN proposals.  For instance,
   although widely deployed mobile network architectures typically come
   with their own network entities and associated protocols, they follow
   the same line of design with respect to managing mobility.  This
   design thinking, which calls for incorporating mobility anchors,
   permeates in the ICN literature too.

   However, employing mobility anchors and tunneling is probably not the
   best way forward in ICN research for mobile networking.
   Fundamentally, this approach is anything but information-centric and
   location-independent.  In addition, as argued in [SEEN], current
   mobility management schemes anchor information retrieval not only at
   a specific network gateway (e.g., home agent in Mobile IP) but also
   at a specific correspondent node due to the end-to-end nature of
   host-centric communication.  However, once a change in the point of
   attachment occurs, information retrieval from the original
   "correspondent node" may no longer be optimal.  This was shown in
   [MANI], for example, where a simple mechanism that triggers the



Kutscher, et al.              Informational                    [Page 18]

RFC 7927                     ICN Challenges                    July 2016


   discovery of new retrieval providers for the same data object,
   following a change in the point of attachment, clearly outperforms a
   tunnel-based approach like Mobile IP in terms of object download
   times.  The challenge here is how to capitalize on location
   information while facilitating the use of ICN primitives, which
   natively support multicast and anycast.

   ICN naming and name resolution, as well as the security features that
   come along, should natively support mobility.  For example, CCN [CCN]
   does not have the restriction of spanning tree routing, so it is able
   to take advantage of multiple interfaces or adapt to the changes
   produced by rapid mobility (i.e., there is no need to bind a layer 3
   address with a layer 2 address).  In fact, client mobility can be
   simplified by allowing requests for new content to normally flow from
   different interfaces or through newly connected points of attachment
   to the network.  However, when the node moving is the (only) content
   source, it appears that more complex network support might be
   necessary, including forwarding updates and cache rebuilding.  A case
   in point is a conversation network service, such as a voice or video
   call between two parties.  The requirements in this case are more
   stringent when support for seamless mobility is required, especially
   when compared to content dissemination that is amenable to buffering.
   Another parameter that needs to be paid attention to is the impact of
   using different wireless access interfaces based on different
   technologies, where the performance and link conditions can vary
   widely depending of numerous factors.

   In host-centric networking, mobility management mechanisms ensure
   optimal handovers and (ideally) seamless transition from one point of
   attachment to another.  In ICN, however, the traditional meaning of
   "point of attachment" no longer applies as communication is not
   restrained by location-based access to data objects.  Therefore, a
   "seamless transition" in ICN ensures that content reception continues
   without any perceptible change from the point of view of the ICN
   application receiving that content.  Moreover, this transition needs
   to be executed in parallel with ICN content identification and
   delivery mechanisms, enabling scenarios such as preparation of the
   content delivery process at the target connectivity point prior to
   the handover (to reduce link switch disturbances).  Finally, these
   mobility aspects can also be tightly coupled with network management
   aspects, in respect to policy enforcement, link control, and other
   parameters necessary for establishing the node's link to the network.

   In summary, the following research challenges for ICN mobility
   management can be derived:

   o  How can mobility management take full advantage of native ICN
      primitives?



Kutscher, et al.              Informational                    [Page 19]

RFC 7927                     ICN Challenges                    July 2016


   o  How do we avoid the need for mobility anchors in a network that by
      design supports multicast, anycast, and location-independent
      information retrieval?

   o  How can content retrieval mechanisms interface with specific link
      operations, such as identifying which links are available for
      certain content?

   o  How can mobility be offered as a service that is only activated
      when the specific user/content/conditions require it?

   o  How can mobility management be coordinated between the node and
      the network for optimization and policing procedures?

   o  How do we ensure that managing mobility does not introduce
      scalability issues in ICN?

   o  How will the name resolution process be affected by rapid
      topological changes when the content source itself is mobile?

4.5.  Wireless Networking



   Today, all layer 2 (L2) wireless network radio access technologies
   are developed with a clear assumption in mind: the waist of the
   protocol stack is IP, and it will be so for the foreseeable future.
   By fixing the protocol stack waist, engineers can answer a large set
   of questions, including how to handle conversational traffic (e.g.,
   voice calls) vs. web traffic, how to support multicast, and so on, in
   a rather straightforward manner.  Broadcast, on the other hand, which
   is inherent in wireless communication, is not fully taken advantage
   of.  On the contrary, researchers are often more concerned about
   introducing mechanisms that ensure that "broadcast storms" do not
   take down a network.  The question of how can broadcast better serve
   ICN needs has yet to be thoroughly investigated.

   Wireless networking is often intertwined with mobility, but this is
   not always the case.  In fact, empirical measurements often indicate
   that many users tend to connect (and remain connected) to a single
   Wi-Fi access point for considerable amounts of time.  A case in
   point, which is frequently cited in different variations in the ICN
   literature, is access to a document repository during a meeting.  For
   instance, in a typical IETF working group meeting, a scribe takes
   notes, which are uploaded to a centralized repository (see Figure 1).
   Subsequently, each meeting participant obtains a copy of the document
   on their own devices for local use, annotation, and sharing with
   colleagues that are not present at the meeting.  Note that in this
   example, there is no node mobility and that it is not important




Kutscher, et al.              Informational                    [Page 20]

RFC 7927                     ICN Challenges                    July 2016


   whether the document with the notes is uploaded in one go at the end
   of the session or in a streaming-like fashion as is typical today
   with online (cloud-based) document processing.

           +---------------------+
           | Document Repository |
           +---------------------+
                     ||
                 (Internet)
                     ||
             +--------------+
             | Access Point |
             +--------------+
            /  |             \
           /   |              \
          /    |               \
     Scribe   Participant 1 ... Participant N

                Figure 1: Document Sharing During a Meeting

   In this scenario, we observe that the same data object bits
   (corresponding to the meeting notes) need to traverse the wireless
   medium at least N+1 times, where N is the number of meeting
   participants obtaining a copy of the notes.  In effect, a broadcast
   medium is shoehorned into N+1 virtual unicast channels.  One could
   argue that wireless local connectivity is inexpensive, but this is
   not the critical factor in this example.  The actual information
   exchange wastes N times the available network capacity, no matter
   what the spectral efficiency (or the economics) underlying the
   wireless technology is.  This waste is a direct result of extending
   the remote access paradigm from wired to wireless communication,
   irrespective of the special characteristics of the latter.

   It goes without saying that an ICN approach that does not take into
   consideration the wireless nature of an interface will waste the same
   amount of resources as a host-centric paradigm.  In-network caching
   at the wireless access point could reduce the amount of data carried
   over the backhaul link, but, if there is no change in the use of the
   wireless medium, the NDO will still be carried over the wireless
   ether N+1 times.  Intelligent caching strategies, replica placement
   cooperation, and so on simply cannot alleviate this.  On the other
   hand, promiscuous interface operation and opportunistic caching would
   maximize wireless network capacity utilization in this example.

   Arguably, if one designs a future wireless access technology with an
   information-centric "layer 3" in mind, many of the design choices
   that are obvious in an all-IP architecture may no longer be valid.




Kutscher, et al.              Informational                    [Page 21]

RFC 7927                     ICN Challenges                    July 2016


   Although this is clearly outside the scope of this document, a few
   research challenges that the wider community may be interested in
   include:

   o  Can we use wireless resources more frugally with the information-
      centric paradigm than what is possible today in all-IP wireless
      networks?

   o  In the context of wireless access, how can we leverage the
      broadcast nature of the medium in an information-centric network?

   o  Would a wireless-oriented ICN protocol stack deliver significant
      performance gains?  How different would it be from a wired-
      oriented ICN protocol stack?

   o  Is it possible that by changing the network paradigm to ICN we
      can, in practice, increase the spectral efficiency (bits/s/Hz) of
      a wireless network beyond what would be possible with today's
      host-centric approaches?  What would be the impact of doing so
      with respect to energy consumption?

   o  Can promiscuous wireless interface operation coupled with
      opportunistic caching increase ICN performance, and if so, by how
      much?

   o  How can a conversational service be supported at least as
      efficiently as today's state-of-the-art wireless networks deliver?

   o  What are the benefits of combining ICN with network coding in
      wireless networks?

   o  How can Multiple-Input Multiple-Output (MIMO) and Coordinated
      Multipoint Transmission (CoMP) be natively combined with ICN
      primitives in future cellular networks?

4.6.  Rate and Congestion Control



   ICN's receiver-driven communication model as described above creates
   new opportunities for transport protocol design, as it does not rely
   solely on end-to-end communication from a sender to a requestor.  A
   requested data object can be accessible in multiple different network
   locations.  A node can thus decide how to utilize multiple sources,
   e.g., by sending parallel requests for the same NDO or by switching
   sources (or next hops) in a suitable schedule for a series of
   requests.






Kutscher, et al.              Informational                    [Page 22]

RFC 7927                     ICN Challenges                    July 2016


   In this model, the requestor would control the data rate by
   regulating its request sending rate and next by performing source/
   next-hop selections.  Specific challenges depend on the specific ICN
   approach, but general challenges for receiver-driven transport
   protocols (or mechanisms, since dedicated protocols might not be
   required) include flow and congestion control, fairness, network
   utilization, stability (of data rates under stable conditions), etc.
   [HRICP] and [CONTUG] describe request rate control protocols and
   corresponding design challenges.

   As mentioned above, the ICN communication paradigm does not depend
   strictly on end-to-end flows, as contents might be received from in-
   network caches.  The traditional concept of a flow is then somewhat
   not valid as sub-flows, or flowlets, might be formed on the fly, when
   fractions of an NDO are transmitted from in-network caches.  For a
   transport-layer protocol, this is challenging, as any measurement
   related to this flow as traditionally done by transport protocols
   such as TCP, can often prove misleading.  For example, false Round-
   Trip Time (RTT) measurements will lead to largely variable average
   and smoothed RTT values, which in turn will trigger false timeout
   expirations.

   Furthermore, out-of-order delivery is expected to be common in a
   scenario where parts of a data object are retrieved from in-network
   caches rather than from the origin server.  Several techniques for
   dealing with out-of-order delivery have been proposed in the past for
   TCP, some of which could potentially be modified and reused in the
   context of ICN.  Further research is needed in this direction though
   to choose the right technique and adjust it according to the
   requirements of the ICN architecture and transport protocol in use.

   ICN offers routers the possibility to aggregate requests and can use
   several paths, meaning that there is no such thing as a (dedicated)
   end-to-end communication path, e.g., a router that receives two
   requests for the same content at the same time only sends one request
   to its neighbor.  The aggregation of requests has a general impact on
   transport protocol design and offers new options for employing per-
   node forwarding strategies and for rethinking in-network resource
   sharing [RESOURCE-POOL].

   Achieving fairness for requestors can be one challenge as it is not
   possible to identify the number of requestors behind one particular
   request.  A second problem related to request aggregation is the
   management of request retransmissions.  Generally, it is assumed that
   a router will not transmit a request if it transmitted an identical
   request recently, and because there is no information about the
   requestor, the router cannot distinguish the initial request from a




Kutscher, et al.              Informational                    [Page 23]

RFC 7927                     ICN Challenges                    July 2016


   client from a retransmission from the same client.  In such a
   situation, routers can adapt their timers to use the best of the
   communication paths.

4.7.  In-Network Caching



   Explicitly named data objects allow for caching at virtually any
   network element, including routers, proxy caches, and end-user
   devices.  Therefore, in-network caching can improve network
   performance by fetching content from nodes that are geographically
   placed closer to the end user.  Several issues that need further
   investigation have been identified with respect to in-network
   caching.  In this section, we list important challenges that relate
   to the properties of the new ubiquitous caching system.

4.7.1.  Cache Placement



   The declining cost of fast memory gives the opportunity to deploy
   caches in network routers and to take advantage of cached NDOs.  We
   identify two approaches to in-network caching, namely, on-path and
   off-path caching.  Both approaches have to consider the issue of
   cache location.  Off-path caching is similar to traditional proxy-
   caching or CDN server placement.  Retrieval of contents from off-path
   caches requires redirection of requests and, therefore, is closely
   related to the Request-to-Cache Routing problem discussed below.
   Off-path caches have to be placed in strategic points within a
   network in order to reduce the redirection delays and the number of
   detour hops to retrieve cached contents.  Previous research on proxy-
   caching and CDN deployment is helpful in this case.

   On the other hand, on-path caching requires less network intervention
   and fits more neatly in ICN.  However, on-path caching requires line-
   speed operation, which places more constraints on the design and
   operation of in-network caching elements.  Furthermore, the gain of
   such a system of on-path in-network caches relies on opportunistic
   cache hits and has therefore been considered of limited benefit,
   given the huge amount of contents hosted in the Internet.  For this
   reason, network operators might initially consider only a limited
   number of network elements to be upgraded to in-network caching
   elements.  The decision on which nodes should be equipped with caches
   is an open issue and might be based, among others, on topological
   criteria or traffic characteristics.  These challenges relate to both
   the Content Placement problem and the Request-to-Cache Routing
   problem discussed below.

   In most cases, however, the driver for the implementation,
   deployment, and operation of in-network caches will be its cost.
   Operating caches at line speed inevitably requires faster memory,



Kutscher, et al.              Informational                    [Page 24]

RFC 7927                     ICN Challenges                    July 2016


   which increases the implementation cost.  Based on the capital to be
   invested, ISPs will need to make strategic decisions on the cache
   placement, which can be driven by several factors, such as avoidance
   of inter-domain/expensive links, centrality of nodes, size of domain
   and the corresponding spatial locality of users, and traffic patterns
   in a specific part of the network (e.g., university vs. business vs.
   fashion district of a city).

4.7.2.  Content Placement: Content-to-Cache Distribution



   Given a number of on-path or off-path in-network caching elements,
   content-to-cache distribution will affect both the dynamics of the
   system, in terms of request redirections (mainly in case of off-path
   caches) and the gain of the system in terms of cache hits.  A
   straightforward approach to content placement is on-path placement of
   contents as they travel from source to destination.  This approach
   reduces the computation and communication overhead of placing content
   within the network but, on the other hand, might reduce the chances
   of hitting cached contents.  This relates to the Request-to-Cache
   Routing problem discussed next.

   Furthermore, the number of replicas held in the system brings up
   resource management issues in terms of cache allocation.  For
   example, continuously replicating data objects in all network
   elements results in redundant copies of the same objects.  The issue
   of redundant replication has been investigated in the past for
   hierarchical web caches.  However, in hierarchical web-caching,
   overlay systems coordination between the data and the control plane
   can guarantee increased performance in terms of cache hits.  Line-
   speed, on-path, in-network caching poses different requirements;
   therefore, new techniques need to be investigated.  In this
   direction, reducing the redundancy of cached copies is a study item.
   However, the issue of coordinated content placement in on-path caches
   remains open.

   The Content-to-Cache Allocation problem relates also to the
   characteristics of the content to be cached.  Popular content might
   need to be placed where it is going to be requested next.
   Furthermore, issues of "expected content popularity" or temporal
   locality need to be taken into account in designing in-network
   caching algorithms in order for some contents to be given priority
   (e.g., popular content vs. one-timers).  The criteria as to which
   contents should be given priority in in-network content caches
   relates also to the business relationships between content providers
   and network operators.  Business model issues will drive some of
   these decisions on content-to-cache distribution, but such issues are
   outside the scope of this note and are not discussed here further.




Kutscher, et al.              Informational                    [Page 25]

RFC 7927                     ICN Challenges                    July 2016


4.7.3.  Request-to-Cache Routing



   In order to take advantage of cached contents, requests have to be
   forwarded to the nodes that cache the corresponding contents.  This
   challenge relates to name-based routing, discussed earlier.  Requests
   should ideally follow the path to the cached NDO.  However,
   instructions as to which content is cached where cannot be broadcast
   throughout the network.  Therefore, the knowledge of an NDO location
   at the time of the request either might not exist or might not be
   accurate (i.e., contents might have been removed by the time a
   request is redirected to a specific node).

   Coordination between the data and the control planes to update
   information of cached contents has been considered, but in this case,
   scalability issues arise.  We therefore have two options.  We either
   have to rely on opportunistic caching, where requests are forwarded
   to a server and in case the NDO is found on the path, then the
   content is fetched from this node instead of the origin server, or we
   employ cache-aware routing techniques.  Cache-aware routing can
   involve either both the control and the data plane or only one of
   them.  Furthermore, cache-aware routing can be done in a domain-wide
   scale or can involve more than one individual Autonomous System (AS).
   In the latter case, business relationships between ASes might need to
   be exploited in order to build a scalable model.

4.7.4.  Staleness Detection of Cached NDOs



   Due to the largely distributed copies of NDOs in in-network caches,
   ICN should be able to provide a staleness verification algorithm that
   provides synchronization of NDOs located at their providers and in-
   network caching points.  Two types of approaches can be considered
   for this problem, namely direct and indirect approaches.

   In the direct approach, each cache looks up certain information in
   the NDO's name, e.g., the timestamp, that directly indicates its
   staleness.  This approach is applicable to some NDOs that come from
   machine-to-machine and Internet of Things scenarios, whose base
   operation relies on obtaining the latest version of that NDO (i.e., a
   soil sensor in a farm providing different continuous parameters that
   are sent to a display or greenhouse regulation system) [FRESHNESS].

   In the indirect approach, each cache consults the publisher of the
   cached NDO about its staleness before serving it.  This approach
   assumes that the NDO includes the publisher information, which can be
   used to reach the publisher.  It is suitable for the NDO whose
   expiration time is difficult to be set in advance, e.g., a web page





Kutscher, et al.              Informational                    [Page 26]

RFC 7927                     ICN Challenges                    July 2016


   that contains the main text (which stays the same ever after) and the
   interactive sections such as comments or ads (which are updated
   irregularly).

   It is often argued that ignoring stale NDOs in caches and simply
   providing new names for updated NDOs might be sufficient rather than
   using a staleness verification algorithm to manage them.  However,
   notifying the new names of updated NDOs to users is not a trivial
   task.  Unless the update is informed to all users at the same time,
   some users would use the old name although they intended to retrieve
   the updated NDO.

   One research challenge is how to design consistency and coherence
   models for caching NDOs along with their revision handling and
   updating protocols in a scalable manner.

4.7.5.  Cache Sharing by Multiple Applications



   When ICN is deployed as a general, application-independent network
   and cache infrastructure, multiple consumers and producers
   (representing different applications) would communicate over the same
   infrastructure.  With universal naming schemes or sufficiently unique
   hash-based identifiers, different application could also share
   identical NDOs in a transparent way.

   Depending on the naming, data integrity, and data origin
   authentication approaches, there may be technical and business
   challenges to share caches across different applications, for
   example, content protection, avoiding cache poisoning, ensuring
   performance isolation, etc.  As ICN research matures, these
   challenges should be addressed more specifically in a dedicated
   document.

4.8.  Network Management



   Managing networks has been a core craft in the IP-based host-centric
   paradigm ever since the technology was introduced in production
   networks.  However, at the onset of IP, management was considered
   primarily as an add-on.  Essential tools that are used daily by
   networkers, such as ping and traceroute, did not become widely
   available until more than a decade or so after IP was first
   introduced.  Management protocols, such as SNMP, also became
   available much later than the original introduction of IP, and many
   still consider them insufficient despite the years of experience we
   have running host-centric networks.  Today, when new networks are
   deployed, network management is considered a key aspect for any
   operator, a major challenge that is directly reflected in higher
   operational cost if not done well.  If we want ICN to be deployed in



Kutscher, et al.              Informational                    [Page 27]

RFC 7927                     ICN Challenges                    July 2016


   infrastructure networks, development of management tools and
   mechanisms must go hand in hand with the rest of the architecture
   design.

   Although defining an FCAPS (Fault, Configuration, Accounting,
   Performance, and Security) [ISOIEC-7498-4] management model for ICN
   is clearly outside the scope of this document, there is a need for
   creating basic tools early on while ICN is still in the design and
   experimentation phases that can evolve over time and help network
   operations centers (NOCs) to define policies, validate that they are
   indeed used in practice, be notified early on about failures, and
   determine and resolve configuration problems.  Authentication,
   Authorization, and Accounting (AAA) as well as performance
   management, from a NOC perspective, will also need to be considered.
   Given the expectations for a large number of nodes and unprecedented
   traffic volumes, automating tasks or even better employing self-
   management mechanisms are preferred.  The main challenge here is that
   all tools we have at our disposal today are node-centric, are end-to-
   end oriented, or assume a packet-stream communication environment.
   Rethinking reachability and operational availability, for example,
   can yield significant insights into how information-centric networks
   will be managed in the future.

   With respect to network management, we see three different aspects.
   First, any operator needs to manage all resources available in the
   network, which can range from node connectivity to network bandwidth
   availability to in-network storage to multi-access support.  In ICN,
   users will also bring into the network significant resources in terms
   of network coverage extension, storage, and processing capabilities.
   Delay Tolerant Networking (DTN) characteristics should also be
   considered to the degree that this is possible (e.g., content
   dissemination through data mules).  Second, given that nodes and
   links are not at the center of an information-centric network,
   network management should capitalize on native ICN mechanisms.  For
   example, in-network storage and name resolution can be used for
   monitoring, while native publish/subscribe functionality can be used
   for triggering notifications.  Finally, management is also at the
   core of network-controlling capabilities by allowing operating
   actions to be mediated and decided, triggering and activating
   networking procedures in an optimized way.  For example, monitoring
   aspects can be conjugated with different management actions in a
   coordinated way, allowing network operations to flow in a concerted
   manner.

   However, the considerations on leveraging intrinsic ICN mechanisms
   and capabilities to support management operations go beyond a simple
   mapping exercise.  In fact, it not only raises a series of challenges
   on its own, but also opens up new possibilities for both ICN and



Kutscher, et al.              Informational                    [Page 28]

RFC 7927                     ICN Challenges                    July 2016


   "network management" as a concept.  For instance, naming mechanisms
   are central to ICN-intrinsic operations, which are used to identify
   and reach content under different aspects (e.g., hierarchically
   structured vs. "flattish" names).  In this way, ICN is decoupled from
   host-centric aspects on which traditional network management schemes
   rely.  As such, questions are raised that can directly be translated
   into challenges for network management capability, such as, for
   example, how to address a node or a network segment in an ICN naming
   paradigm, how to identify which node is connected "where", how to be
   aware of the node capabilities (i.e., high or low-powered machine-to-
   machine (M2M) node), and if there is a host-centric protocol running
   where the management process can also leverage.

   But, on the other hand, these same inherent ICN characteristics also
   allow us to look into network management through a new perspective.
   By centering its operations around NDOs, one can conceive new
   management operations addressing, for example, per-content management
   or access control, as well as analyzing performance per NDO instead
   of per link or node.  Moreover, such considerations can also be used
   to manage operational aspects of ICN mechanisms themselves.  For
   example, [NDN-MGMT] reutilizes inherent content-centric capabilities
   of CCN to manage optimal link connectivity for nodes, in concert with
   a network optimization process.  Conversely, how these content-
   centric aspects can otherwise influence and impact management in
   other areas (e.g., security and resilience) is also important, as
   exemplified in [CCN-ACCESS], where access control mechanisms are
   integrated into a prototype of the [PURSUIT] architecture.

   The set of core research challenges for ICN management includes:

   o  Management and control of NDO reception at the requestor

   o  Coordination of management information exchange and control
      between ICN nodes and ICN network control points

   o  Identification of management and controlling actions and items
      through information naming

   o  Relationship between NDOs and host entities identification, i.e.,
      how to identify a particular link, interface, or flow that needs
      to be managed

4.9.  ICN Applications



   ICN can be applied to different application domains and is expected
   to provide benefits for application developers by providing a more
   suitable interface for application developers (in addition to the




Kutscher, et al.              Informational                    [Page 29]

RFC 7927                     ICN Challenges                    July 2016


   other ICN benefits described above).  [RFC7476] provides an overview
   of relevant application domains at large.  This section discusses
   opportunities and challenges for selected application types.

4.9.1.  Web Applications



   Intuitively, the ICN request/response communication style seems to be
   directly mappable to web communication over HTTP.  NDO names could be
   the equivalent of URIs in today's web, proprietary and transparent
   caching could be obsoleted by ICN in-network caching, and developers
   could directly use an ICN request/response API to build applications.

   Research efforts such as [ICN2014-WEB-NDN] have analyzed real-world
   web applications and ways to implement them in ICN.  The most
   significant insight is that REST-style (Representational State
   Transfer) web communication relies heavily on transmitting user/
   application context information in HTTP GET requests, which would
   have to be mapped to corresponding ICN messages.  The challenge in
   ICN would be how to exactly achieve that mapping.  This could be done
   to some degree by extending name formats or by extending message
   structure to include cookies and similar context information.  The
   design decisions would need to consider overhead in routers (for
   example, if larger GET/Interest messages would have to be stored in
   corresponding tables on routers).

   Other challenges include the ability to return different results
   based on requestor-specific processing in the presence of immutable
   objects (and name-object bindings) in ICN and the ability for
   efficient bidirectional communication, which would require some
   mechanism to name and reach requestor applications.

4.9.2.  Video Streaming and Download



   One of ICN's prime application areas is video streaming and download
   where accessing named data, object-level security, and in-network
   storage can fulfill requirements for both video streaming and
   download.  The applicability and benefits of ICN to video has been
   demonstrated by several prototype developments
   [ICN2014-AHLGREN-VIDEO-DEMO].

   [VIDEO-STREAMING] discusses the opportunities and challenges of
   implementing today's video services such as DASH-based (Dynamic
   Adaptive Streaming over HTTP) streaming and download over ICN,
   considering performance requirements, relationship to peer-to-peer
   live streaming, IPTV, and Digital Rights Management (DRM).






Kutscher, et al.              Informational                    [Page 30]

RFC 7927                     ICN Challenges                    July 2016


   In addition to just porting today's video application from a host-
   centric paradigm to ICN, there are also promising opportunities to
   leverage the ICN network services for redesigning and thus
   significantly enhancing video access and distribution
   [ICNRG-2015-01-WESTPHAL].  For example, ICN store and forward could
   be leveraged for rate adaptation to achieve maximum throughput and
   optimal Quality of Experience (QoE) in scenarios with varying link
   properties, if capacity information is fed back to rate selection
   algorithms at senders.  Other optimizations such as more aggressive
   prefetching could be performed in the network by leveraging
   visibility of chunk NDO names and NDO metadata in the network.
   Moreover, multi-source rate adaptation in combination with network
   coding could enable better QoE, for example, in multi-interface/
   access scenarios where multiple paths from client to upstream caches
   exist [RFC7476].

4.9.3.  Internet of Things



   The essence of ICN lies in the name-based routing that enables users
   to retrieve NDOs by the names regardless of their locations.  By
   definition, ICN is well suited for IoT applications, where users
   consume data generated from IoTs without maintaining secure
   connections to them.  The basic request/response style APIs of ICN
   enable developers to build IoT applications in a simple and fast
   manner.

   Ongoing efforts such as [ICN-FOR-IOT], [ICN-ARCH], and
   [ICN2014-NDNWILD] have addressed the requirements and challenges of
   ICN for IoT.  For instance, many IoT applications depend on a PUSH
   model where data transmission is initiated by the publisher, so they
   can support various real-time applications (emergency alarm, etc.).
   However, ICN does not support the PUSH model in a native manner due
   to its inherent receiver-driven data transmission mechanism.  The
   challenge would be how to efficiently support the PUSH model in ICN,
   so it provides publish/subscribe-style APIs for IoT application
   developers.  This could be done by introducing other types of
   identifiers such as a device identifier or by extending the current
   request/response communication style, which may result in heavy
   overhead in ICN routers.

   Moreover, key characteristics of the ICN underlying operation also
   impact important aspects of IoT, such as the caching in content
   storage of network forwarding entities.  This allows the
   simplification of ICN-based IoT application development.  Since the
   network is able to act on named content, generic names provide a way
   to address content independently of the underlying device (and
   access) technology, and bandwidth consumption is optimized due to the
   availability of cached content.  However, these aspects raise



Kutscher, et al.              Informational                    [Page 31]

RFC 7927                     ICN Challenges                    July 2016


   challenges themselves concerning the freshness of the information
   received from the cache in contrast to the last value generated by a
   sensor, as well as pushing content to specific nodes (e.g., for
   controlling them), which requires individual addressing for
   identification.  In addition, due to the heterogeneous nature of IoT
   nodes, their processing capabilities might not be able to handle the
   necessary content signing verification procedures.

5.  Security Considerations



   This document does not impact the security of the Internet.  Security
   questions related to ICN are discussed in Section 4.2.

6.  Informative References



   [ACCESS-CTL-DEL]
              Fotiou, N., Marias, G., and G. Polyzos, "Access control
              enforcement delegation for information-centric networking
              architectures", Proceedings of the second edition of the
              ICN workshop on Information-centric networking (ICN
              '12) Helsinki, Finland, DOI 10.1145/2342488.2342507, 2012.

   [BACKSCATTER]
              Waehlisch, M., Schmidt, TC., and M. Vahlenkamp,
              "Backscatter from the Data Plane - Threats to Stability
              and Security in Information-Centric Network
              Infrastructure", Computer Networks Vol 57, No. 16, pp.
              3192-3206, DOI 10.1016/j.comnet.2013.07.009, November
              2013.

   [BREADCRUMBS]
              Rosensweig, E. and J. Kurose, "Breadcrumbs: Efficient,
              Best-Effort Content Location in Cache Networks",
              In Proceedings of the IEEE INFOCOM 2009,
              DOI 10.1109/INFCOM.2009.5062201, April 2009.

   [CCN]      Jacobson, V., Smetters, D., Thornton, J., Plass, M.,
              Briggs, N., and R. Braynard, "Networking Named Content",
              CoNEXT 2009, DOI 10.1145/1658939.1658941, December 2009.

   [CCN-ACCESS]
              Fotiou, N., Marias, G., and G. Polyzos, "Access control
              enforcement delegation for information-centric networking
              architectures", In Proceedings of the second edition of
              the ICN workshop on Information-centric networking (ICN
              '12), ACM, New York, NY, USA, 85-90,
              DOI 10.1145/2342488.2342507, 2012.




Kutscher, et al.              Informational                    [Page 32]

RFC 7927                     ICN Challenges                    July 2016


   [CHAUM]    Chaum, D. and E. van Heijst, "Group signatures",
              In Proceedings of EUROCRYPT, DOI 10.1007/3-540-46416-6_22,
              1991.

   [COMPACT]  Cowen, L., "Compact routing with minimum stretch",
              In Journal of Algorithms, vol. 38, pp. 170-183,
              DOI 10.1006/jagm.2000.1134, 2001.

   [CONTUG]   Arianfar, S., Nikander, P., Eggert, L., Ott, J., and W.
              Wong, "ConTug: A Receiver-Driven Transport Protocol for
              Content-Centric Networks", Technical Report Aalto
              University Comnet, 2011.

   [DONA]     Koponen, T., Ermolinskiy, A., Chawla, M., Kim, K., gon
              Chun, B., and S. Shenker, "A Data-Oriented (and Beyond)
              Network Architecture", In Proceedings of SIGCOMM 2007,
              DOI 10.1145/1282427.1282402, August 2007.

   [ENCRYPTION-AC]
              Kurihara, J., Uzun, E., and C. Wood, "An Encryption-Based
              Access Control Framework for Content-Centric Networking",
              IFIP Networking 2015, Toulouse, France,
              DOI 10.1109/IFIPNetworking.2015.7145300, September 2015.

   [FRESHNESS]
              Quevedo, J., Corujo, D., and R. Aguiar, "Consumer Driven
              Information Freshness Approach for Content Centric
              Networking", IEEE INFOCOM Workshop on Name-Oriented
              Mobility Toronto, Canada,
              DOI 10.1109/INFCOMW.2014.6849279, May 2014.

   [GREEDY]   Papadopoulos, F., Krioukov, D., Boguna, M., and A. Vahdat,
              "Greedy forwarding in dynamic scale-free networks embedded
              in hyperbolic metric spaces", In Proceedings of the IEEE
              INFOCOM, San Diego, USA, DOI 10.1109/INFCOM.2010.5462131,
              2010.

   [HRICP]    Carofiglio, G., Gallo, M., and L. Muscariello, "Joint hop-
              by-hop and receiver-driven interest control protocol for
              content-centric networks", In Proceedings of ACM SIGCOMM
              ICN 2012, DOI 10.1145/2342488.2342497, 2012.

   [ICN-ARCH] Zhang, Y., Raychadhuri, D., Grieco, L., Baccelli, E.,
              Burke, J., Ravindran, R., Ed., and G. Wang, "ICN based
              Architecture for IoT - Requirements and Challenges", Work
              in Progress, draft-zhang-iot-icn-challenges-02, August
              2015.




Kutscher, et al.              Informational                    [Page 33]

RFC 7927                     ICN Challenges                    July 2016


   [ICN-FOR-IOT]
              Lindgren, A., Ben Abdesslem, F., Ahlgren, B., Schelen, O.,
              and A. Malik, "Applicability and Tradeoffs of Information-
              Centric Networking for Efficient IoT", Work in Progress,
              draft-lindgren-icnrg-efficientiot-03, July 2015.

   [ICN2014-AHLGREN-VIDEO-DEMO]
              Ahlgren, B., Jonasson, A., and B. Ohlman, "Demo Overview:
              HTTP Live Streaming over NetInf Transport", ACM SIGCOMM
              Information-Centric Networking Conference Paris, France,
              DOI 10.1145/2660129.2660136, September 2014.

   [ICN2014-NDNWILD]
              Baccelli, E., Mehlis, C., Hahm, O., Schmidt, T., and M.
              Waehlisch, "Information Centric Networking in the IoT:
              Experiments with NDN in the Wild", ACM SIGCOMM
              Information-Centric Networking Conference Paris, France,
              DOI 10.1145/2660129.2660144, September 2014.

   [ICN2014-WEB-NDN]
              Moiseenko, I., Stapp, M., and D. Oran, "Communication
              Patterns for Web Interaction in Named Data Networking",
              ACM SIGCOMM Information-Centric Networking
              Conference Paris, France, DOI 10.1145/2660129.2660152,
              September 2014.

   [ICNNAMING]
              Ghodsi, A., Koponen, T., Rajahalme, J., Sarolahti, P., and
              S. Shenker, "Naming in Content-Oriented Architectures",
              In Proceedings ACM SIGCOMM Workshop on Information-Centric
              Networking (ICN), DOI 10.1145/2018584.2018586, 2011.

   [ICNRG-2015-01-WESTPHAL]
              Westphal, C., "Video over ICN", IRTF ICNRG
              Meeting Cambridge, Massachusetts, USA, January 2015,
              <http://www.ietf.org/proceedings/interim/2015/01/13/icnrg/
              slides/slides-interim-2015-icnrg-1-0.pptx>.

   [ICNSURVEY]
              Ahlgren, B., Dannewitz, C., Imbrenda, C., Kutscher, D.,
              and B. Ohlman, "A Survey of Information-Centric
              Networking", In Communications Magazine, IEEE, vol. 50,
              no. 7, pp. 26-36, DOI 10.1109/MCOM.2012.6231276, 2012.








Kutscher, et al.              Informational                    [Page 34]

RFC 7927                     ICN Challenges                    July 2016


   [ISOIEC-7498-4]
              ISO, "Information Processing Systems -- Open Systems
              Interconnection -- Basic Reference Model -- Part 4:
              Management Framework", November 1989,
              <http://standards.iso.org/ittf/PubliclyAvailableStandards/
              s014258_ISO_IEC_7498-4_1989(E).zip>.

   [MANI]     Pentikousis, K. and T. Rautio, "A multiaccess Network of
              Information", WoWMoM 2010 IEEE,
              DOI 10.1109/WOWMOM.2010.5534922, June 2010.

   [MDHT]     D'Ambrosio, M., Dannewitz, C., Karl, H., and V.
              Vercellone, "MDHT: A hierarchical name resolution service
              for information-centric networks", ACM SIGCOMM workshop on
              Information-centric networking Toronto, Canada,
              DOI 10.1145/2018584.2018587, August 2011.

   [MMIN]     Pentikousis, K. and P. Bertin, "Mobility management in
              infrastructure networks", Internet Computing, IEEE vol.
              17, no. 5, pp. 74-79, DOI 10.1109/MIC.2013.98, October
              2013.

   [NDN-CTL-SHARING]
              Yu, Y., "Controlled Sharing of Sensitive Content", IRTF
              ICNRG Meeting San Francisco, USA, October 2015,
              <https://www.ietf.org/proceedings/interim/2015/10/03/
              icnrg/slides/slides-interim-2015-icnrg-4-8.pdf>.

   [NDN-MGMT] Corujo, D., Aguiar, R., Vidal, I., and J. Garcia-Reinoso,
              "A named data networking flexible framework for management
              communications", Communications Magazine, IEEE vol. 50,
              no. 12, pp. 36-43, DOI 10.1109/MCOM.2012.6384449, December
              2012.

   [PURSUIT]  Fotiou et al., N., "Developing Information Networking
              Further: From PSIRP to PURSUIT", In Proceedings of Proc.
              BROADNETS. ICST, DOI 10.1007/978-3-642-30376-0_1, 2010.

   [RANDOM]   Gkantsidis, C., Mihail, M., and A. Saberi, "Random walks
              in peer-to-peer networks: algorithms and evaluation",
              In Perform. Eval., vol. 63, pp. 241-263,
              DOI 10.1016/j.peva.2005.01.002, 2006.

   [RESOURCE-POOL]
              Psaras, I., Saino, L., and G. Pavlou, "Revisiting Resource
              Pooling: The case of In-Network Resource Sharing", ACM
              HotNets Los Angeles, USA, DOI 10.1145/2670518.2673875,
              October 2014.



Kutscher, et al.              Informational                    [Page 35]

RFC 7927                     ICN Challenges                    July 2016


   [RFC2002]  Perkins, C., Ed., "IP Mobility Support", RFC 2002,
              DOI 10.17487/RFC2002, October 1996,
              <http://www.rfc-editor.org/info/rfc2002>.

   [RFC4838]  Cerf, V., Burleigh, S., Hooke, A., Torgerson, L., Durst,
              R., Scott, K., Fall, K., and H. Weiss, "Delay-Tolerant
              Networking Architecture", RFC 4838, DOI 10.17487/RFC4838,
              April 2007, <http://www.rfc-editor.org/info/rfc4838>.

   [RFC5246]  Dierks, T. and E. Rescorla, "The Transport Layer Security
              (TLS) Protocol Version 1.2", RFC 5246,
              DOI 10.17487/RFC5246, August 2008,
              <http://www.rfc-editor.org/info/rfc5246>.

   [RFC5280]  Cooper, D., Santesson, S., Farrell, S., Boeyen, S.,
              Housley, R., and W. Polk, "Internet X.509 Public Key
              Infrastructure Certificate and Certificate Revocation List
              (CRL) Profile", RFC 5280, DOI 10.17487/RFC5280, May 2008,
              <http://www.rfc-editor.org/info/rfc5280>.

   [RFC5944]  Perkins, C., Ed., "IP Mobility Support for IPv4, Revised",
              RFC 5944, DOI 10.17487/RFC5944, November 2010,
              <http://www.rfc-editor.org/info/rfc5944>.

   [RFC6920]  Farrell, S., Kutscher, D., Dannewitz, C., Ohlman, B.,
              Keranen, A., and P. Hallam-Baker, "Naming Things with
              Hashes", RFC 6920, DOI 10.17487/RFC6920, April 2013,
              <http://www.rfc-editor.org/info/rfc6920>.

   [RFC7476]  Pentikousis, K., Ed., Ohlman, B., Corujo, D., Boggia, G.,
              Tyson, G., Davies, E., Molinaro, A., and S. Eum,
              "Information-Centric Networking: Baseline Scenarios",
              RFC 7476, DOI 10.17487/RFC7476, March 2015,
              <http://www.rfc-editor.org/info/rfc7476>.

   [RFC7696]  Housley, R., "Guidelines for Cryptographic Algorithm
              Agility and Selecting Mandatory-to-Implement Algorithms",
              BCP 201, RFC 7696, DOI 10.17487/RFC7696, November 2015,
              <http://www.rfc-editor.org/info/rfc7696>.

   [SEEN]     Pentikousis, K., "In search of energy-efficient mobile
              networking", Communications Magazine, IEEE vol. 48 no. 1,
              pp. 95-103, DOI 10.1109/MCOM.2010.5394036, January 2010.








Kutscher, et al.              Informational                    [Page 36]

RFC 7927                     ICN Challenges                    July 2016


   [VIDEO-STREAMING]
              Westphal, C., Ed., Lederer, S., Posch, D., Timmerer, C.,
              Azgin, A., Liu, S., Mueller, C., Detti, A., Corujo, D.,
              Wang, J., Montpetit, M., Murray, N., Azgin, A., and S.
              Liu, "Adaptive Video Streaming over ICN", Work in
              Progress, draft-irtf-icnrg-videostreaming-08, April 2016.

Acknowledgments

   The authors would like to thank Georgios Karagiannis for providing
   suggestions on QoS research challenges, Dimitri Papadimitriou for
   feedback on the routing section, and Joerg Ott and Stephen Farrell
   for reviewing the whole document.

Authors' Addresses



   Dirk Kutscher (editor)
   NEC
   Kurfuersten-Anlage 36
   Heidelberg
   Germany

   Email: kutscher@neclab.eu


   Suyong Eum
   Osaka University, School of Information Science and Technology
   1-5 Yamadaoka, Suita
   Osaka  565-0871
   Japan

   Phone: +81-6-6879-4571
   Email: suyong@ist.osaka-u.ac.jp


   Kostas Pentikousis
   Travelping
   Koernerstr. 7-10
   Berlin  10785
   Germany

   Email: k.pentikousis@travelping.com









Kutscher, et al.              Informational                    [Page 37]

RFC 7927                     ICN Challenges                    July 2016


   Ioannis Psaras
   University College London, Dept. of E.E.  Eng.
   Torrington Place
   London  WC1E 7JE
   United Kingdom

   Email: i.psaras@ucl.ac.uk


   Daniel Corujo
   Universidade de Aveiro
   Instituto de Telecomunicacoes, Campus Universitario de Santiago
   Aveiro  P-3810-193
   Portugal

   Email: dcorujo@av.it.pt


   Damien Saucez
   INRIA
   2004 route des Lucioles - BP 93
   Sophia Antipolis  06902 Cedex
   France

   Email: damien.saucez@inria.fr


   Thomas C. Schmidt
   HAW Hamburg
   Berliner Tor 7
   Hamburg  20099
   Germany

   Email: t.schmidt@haw-hamburg.de


   Matthias Waehlisch
   FU Berlin
   Takustr. 9
   Berlin  14195
   Germany

   Email: waehlisch@ieee.org








Kutscher, et al.              Informational                    [Page 38]