|Internet-Draft||RAW Architecture||April 2020|
|Thubert & Papadopoulos||Expires 3 October 2020||[Page]|
- Intended Status:
Reliable and Available Wireless Architecture
Due to uncontrolled interferences, including the self-induced multipath fading, deterministic networking can only be approached on wireless links. The radio conditions may change -way- faster than a centralized routing can adapt and reprogram, in particular when the controller is distant and connectivity is slow and limited. RAW separates the routing time scale at which a complex path is recomputed from the forwarding time scale at which the forwarding decision is taken for an individual packet. RAW operates at the forwarding time scale. The RAW problem is to decide, within the redundant solutions that are proposed by the routing, which will be used for each individual packet to provide a DetNet service while minimizing the waste of resources.¶
Status of This Memo
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.¶
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at https://datatracker.ietf.org/drafts/current/.¶
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress."¶
This Internet-Draft will expire on 3 October 2020.¶
Copyright (c) 2020 IETF Trust and the persons identified as the document authors. All rights reserved.¶
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.¶
Bringing determinism in a packet network means eliminating the statistical effects of multiplexing that result in probabilistic jitter and loss. This can be approached with a tight control of the physical resources to maintain the amount of traffic within a budgetted volume of data per unit of time that fits the physical capabilities of the underlying technology, and the use of time-shared resources (bandwidth and buffers) per circuit, and/or by shaping and/or scheduling the packets at every hop.¶
Wireless networks operate on a shared medium where uncontrolled interference, including the self-induced multipath fading, adds another dimension to the statistical effects that affect the delivery. Scheduling transmissions can alleviate those effects by leveraging diversity in the spatial, time, code, and frequency domains, and provide a Reliable and Available service while preserving energy and optimizing the use of the shared spectrum.¶
Deterministic Networking is an attempt to mostly eliminate packet loss for a committed bandwidth with a guaranteed worst-case end-to-end latency, even when co-existing with best-effort traffic in a shared network. This innovation is enabled by recent developments in technologies including IEEE 802.1 TSN (for Ethernet LANs) and IETF DetNet (for wired IP networks). It is getting traction in various industries including manufacturing, online gaming, professional A/V, cellular radio and others, making possible many cost and performance optimizations.¶
The "Deterministic Networking Architecture" [RFC8655] is composed of three planes: the Application (User) Plane, the Controller Plane, and the Network Plane. Reliable and Available Wireless (RAW) extends DetNet to focus on issues that are mostly a co"ern on wireless links, and inherits the architecture and the planes. A RAW Network Plane is thus a Network Plane inherited by RAW from DetNet, composed of one or multiple hops of homogeneous or heterogeneous technologies, e.g. a Wi-Fi6 Mesh or one-hop CBRS access links federated by a 5G backhaul.¶
RAW networking aims at providing highly available and reliable end-to-end performances in a network with scheduled wireless segments. Uncontrolled interference and transmission obstacles may impede the transmission, and techniques such as beamforming with Multi-User MIMO can only alleviate some of those issues, so the term "deterministic" is usually not associated with short range radios, in particular in the ISM band. This uncertainty places limits to the amount of traffic that can be transmitted on a link while conforming to a RAW Service Level Agreement (SLA) that may vary rapidly.¶
The wireless and wired media are fundamentally different at the physical level, and while the generic "Deterministic Networking Problem Statement" [RFC8557] applies to both the wired and the wireless media, the methods to achieve RAW must extend those used to support time-sensitive networking over wires, as a RAW solution has to address less consistent transmissions, energy conservation and shared spectrum efficiency.¶
The development of RAW technologies has been lagging behind deterministic efforts for wired systems both at the IEEE and the IETF. But recent efforts at the IEEE and 3GPP indicate that wireless is finally catching up at the lower layer and that it is now possible for the IETF to extend DetNet for wireless segments that are capable of scheduled wireless transmissions.¶
The intent for RAW is to provide DetNet elements that are specialized for short range radios. From this inheritance, RAW stays agnostic to the radio layer underneath though the capability to schedule transmissions is assumed. How the PHY is programmed to do so, and whether the radio is single-hop or meshed, are unknown at the IP layer and not part of the RAW abstraction.¶
Still, in order to focus on real-worlds issues and assert the feasibility of the proposed capabilities, RAW will focus on selected technologies that can be scheduled at the lower layers: IEEE Std. 802.15.4 timeslotted channel hopping (TSCH), 3GPP 5G ultra-reliable low latency communications (URLLC), IEEE 802.11ax/be where 802.11be is extreme high throughput (EHT), and L-band Digital Aeronautical Communications System (LDACS). See [RAW-TECHNOS] for more.¶
The establishment of a path is not in-scope for RAW. It may be the product of a centralized Controller Plane as described for DetNet. As opposed to wired networks, the action of installing a path over a set of wireless links may be very slow relative to the speed at which the radio conditions vary, and it makes sense in the wireless case to provide redundant forwarding solutions along a complex path and to leave it to the Network Plane to select which of those forwarding solutions are to be used for a given packet based on the current conditions.¶
RAW distinguishes the longer time scale at which routes are computed from the the shorter forwarding time scale where per-packet decisions are made. RAW operates at the forwarding time scale on one DetNet flow over one path that is preestablished and installed by means outside of the scope of RAW. The scope of the RAW WG comprises Network plane protocol elements such as OAM and in-band control to improve the RAW operation at the Service and at the forwarding sub-layers, e.g., controlling whether to use packet replication, Hybrid ARQ and coding, with a constraint to limit the use of redundancy when it is really needed, e.g., when a spike of loss is observed. This is discussed in more details in Section 4.3 and the next sections.¶
RAW defines the following terms:¶
- Packet (hybrid) ARQ, Replication, Elimination and Ordering. PAREO is a superset Of DetNet's PREOF that includes radio-specific techniques such as short range broadcast, MUMIMO, constructive interference and overhearing, which can be leveraged separately or combined to increase the reliability.¶
- In the context of RAW, a link flaps when the wireless connectivity is interrupted for short transient times, typically of a subsecond duration.¶
RAW reuses terminology defined for DetNet in the "Deterministic Networking Architecture" [RFC8655], e.g., PREOF for Packet Replication, Elimination and Ordering Functions.¶
RAW also reuses terminology defined for 6TiSCH in [6TiSCH-ARCH] such as the term Track. 6TiSCH defined a Track as a complex path with associated PAREO operations.¶
This document reuses terms that are well-defined in the context of automation to networking and packet delivery, in particular for reliability and availability. In the context of the RAW work, they are defined as follows:¶
- Reliability is a measure of the probability that an item will perform its intended function for a specified interval under stated conditions. For RAW, the service that is expected is delivery within a bounded latency and a failure is when the packet is either lost or delivered too late. RAW expresses reliability in terms of Mean Time Between Failure (MTBF) and Maximum Consecutive Failures (MCF).¶
- Availability is a measure of the relative amount of time where a path operates in stated condition, in other words (uptime)/(uptime+downtime). Because a serial wireless path may not be good enough to provide the required availability, and even 2 parallel paths may not be over a longer period of time, the RAW availability implies a path that is a lot more complex than what DetNet typically envisages (a Track).¶
3. Use Cases and Requirements Served
[RFC8578] presents a number of wireless use cases including Wireless for Industrial Applications, Pro-Audio and SmartGrid. [RAW-USE-CASES] adds a number of use cases that demonstrate the need for RAW capabilities in new use cases including Pro-Gaming and drones.¶
The use cases can be abstracted in two families, radio access protection and Track protection in a wireless mesh.¶
3.1. Radio Access Protection
In order to maintain the committed reliability at all times, a wireless station may use more than one Radio Access Network (RAN) in parallel.¶
The RANs may be heterogeneous, e.g., 5G [I-D.farkas-5g-raw] and Wi-Fi [RAW-TECHNOS] for high-speed communication, in which case a Layer-3 abstraction becomes useful to select which of the RANs are used at a particular point of time, and the amount of traffic that is distributed over each RAN.¶
The idea is that the rest of the path to the destination(s) is protected separately (e.g., uses non-congruent paths) and/or is a lot more reliable, e.g., wired. In that case, RAW observes reliability of the path through each of the RANs but only acts on the first hop.¶
3.2. Track Protection in a Wireless Mesh
4. RAW Architecture Considerations
4.1. Related Work at The IETF
RAW intersects with protocols or practices in development at the IETF as follows:¶
- The Dynamic Link Exchange Protocol (DLEP) [RFC8175] from [MANET] can be leveraged at each hop to derive generic radio metrics (e.g., based on LQI, RSSI, queueing delays and ETX) on individual hops.¶
- Operations, Administration and Maintenance (OAM) work at [DetNet] such as [DetNet-IP-OAM] for the case of the IP Data Plane observes the state of DetNet paths, typically MPLS and IPv6 pseudowires [DetNet-DP-FW], in the direction of the traffic. RAW needs feedback that flows on the reverse path and gathers instantaneous values from the radio receivers at each hop to inform back the source and replicating relays so they can make optimized forwarding decisions. The work named ICAN may be related as well.¶
- [BFD] detect faults in the path between an ingress and an egress forwarding engines, but is unaware of the complexity of a path with replication, and expects bidirectionality. BFD considers delivery as success whereas with RAW the bounded latency can be as important as the delivery itself.¶
- [SPRING] and [BIER] define in-band signaling that influences the routing when decided at the head-end on the path. There's already one RAW-related draft at BIER [BIER-PREF] more may follow. RAW will need new in-band signaling when the decision is distributed, e.g., required chances of reliable delivery to destination within latency. This signaling enables relays to tune retries and replication to meet the required SLA.¶
- [CCAMP] defines protocol-independent metrics and parameters (measurement attributes) for describing links and paths that are required for routing and signaling in technology-specific networks. RAW would be a source of requirements for CCAMP to define metrics that are significant to the focus radios.¶
A prerequisite to the RAW work is that an end-to-end routing function computes a complex sub-topology along which forwarding can happen between a source and one or more destinations. For 6TiSCH, this is a Track. The concept of Track is specified in the 6TiSCH Architecture [6TiSCH-ARCH]. Tracks provide a high degree of redundancy and diversity and enable DetNet PREOF, end-to-end network coding, and possibly radio-specific abstracted techniques such as ARQ, overhearing, frequency diversity, time slotting, and possibly others.¶
How the routing operation computes the Track is out of scope for RAW. The scope of the RAW operation is one Track, and the goal of the RAW operation is to optimize the use of the Track at the forwarding timescale to maintain the expected service while optimizing the usage of constrained resources such as energy and spectrum.¶
Another prerequisite is that an IP link can be established over the radio with some guarantees in terms of service reliability, e.g., it can be relied upon to transmit a packet within a bounded latency and provides a guaranteed BER/PDR outside rare but existing transient outage windows that can last from split seconds to minutes. The radio layer can be programmed with abstract parameters, and can return an abstract view of the state of the Link to help forwarding decision (think DLEP from MANET). In the layered approach, how the radio manages its PHY layer is out of control and out of scope. Whether it is single hop or meshed is also unknown and out of scope.¶
4.3. Routing Time Scale vs. Forwarding Time Scale
With DetNet, the end-to-end routing can be centralized and can reside outside the network. In wireless, and in particular in a wireless mesh, the path to the controller that performs the route computation and maintenance expensive in terms of critical resources such as air time and energy.¶
Reaching to the routing computation can also be slow in regards to the speed of events that affect the forwarding operation at the radio layer. Due to the cost and latency to perform a route computation, the controller plane is not expected to be sensitive/reactive to transient changes. The abstraction of a link at the routing level is expected to use statistical operational metrics that aggregate the behavior of a link over long periods of time, and represent its availability as shades of gray as opposed to either up or down.¶
In the case of wireless, the changes that affect the forwarding decision can happen frequently and often for short durations, e.g., a mobile object moves between a transmitter and a receiver, and will cancel the line of sight transmission for a few seconds, or a radar measures the depth of a pool and interferes on a particular channel for a split second.¶
There is thus a desire to separate the long term computation of the route and the short term forwarding decision. In such a model, the routing operation computes a complex Track that enables multiple Non-Equal Cost Multi-Path (N-ECMP) forwarding solutions, and leaves it to the forwarding plane to make the per-packet decision of which of these possibilities should be used.¶
In the case of wires, the concept is known in traffic engineering where an alternate path can be used upon the detection of a failure in the main path, e.g., using OAM in MPLS-TP or BFD over a collection of SD-WAN tunnels. RAW formalizes a forwarding time scale that is an order(s) of magnitude shorter than the controler plane routing time scale, and separates the protocols and metrics that are used at both scales. Routing can operate on long term statistics such as delivery ratio over minutes to hours, but as a first approximation can ignore flapping. On the other hand, the RAW forwarding decision is made at packet speed, and uses information that must be pertinent at the present time for the current transmission.¶
4.4. Track Source-Routed vs. Distributed RAW Decision
Within a large routed topology, the route-over mesh operation builds a particular complex Track with one source and one or more destinations; within the Track, packets may follow different paths and may be subject to RAW forwarding operations that include replication, elimination, retries, overhearing and reordering.¶
The RAW forwarding decisions include the selection of points of replication and elimination, how many retries can take place, and a limit of validity for the packet beyond which the packet should be destroyed rather than forwarded uselessly further down the Track.¶
The decision to apply the RAW techniques must be done quickly, and depends on a very recent and precise knowledge of the forwarding conditions within the complex Track. There is a need for an observation method to provide the RAW forwarding plane with the specific knowledge of the state of the Track for the type of flow of interest (e.g., for a QoS level of interest). To observe the whole Track in quasi real time, RAW will consider existing tools such as L2-triggers, DLEP, BFD and in-band and out-of-band OAM.¶
One possible way of making the RAW forwarding decisions is to make them all at the ingress and express them in-band in the packet, which requires new loose or strict Hop-by-hop signaling. To control the RAW forwarding operation along a Track for the individual packets, RAW may leverage and extend known techniques such as DetNet tagging, Segment Routing (SRv6) or BIER-TE such as done with [BIER-PREF].¶
An alternate way is to enable each forwarding node to make the RAW forwarding decisions for a packet on its own, based on its knowledge of the expectation (timeliness and reliability) for that packet and a recent observation of the rest of the way across the possible paths within the Track. Information about the service should be placed in the packet and matched with the forwarding node's capabilities and policies.¶
In either case, a per-flow state is installed in all intermediate nodes to recognize the flow and determine the forwarding policy to be applied.¶
5. RAW Architecture Components
5.1. Reliability and Availability
5.2. Wireless Tracks
The "6TiSCH Architecture" [6TiSCH-ARCH] introduces the concept of Track a a possibly complex path with the PAREO functions operated within.¶
A simple track is composed of a direct sequence of reserved hops to ensure the transmission of a single packet from a source node to a destination node across a multihop path.¶
A Complex Track is designed as a directed acyclic graph from a source node towards a destination node to support multi-path forwarding, as introduced in "6TiSCH Architecture" [6TiSCH-ARCH]. By employing Packet Replication and Elimination (PRE) functions [RFC8655], several paths may be computed, and these paths may be more or less independent. For example, a complex Track may branch off and rejoin over non-congruent paths (branches).¶
Some more details for Deterministic Network PRE techniques are presented in the following Section.¶
5.3. PAREO Functions
In a nutshell, Packet Replication and Elimination (PRE) establishes several paths in a network to provide redundancy and parallel transmissions to bound the end-to-end delay to traverse the network. Optionally, promiscuous listening between paths is possible, such that the nodes on one path may overhear transmissions along the other path. Considering the scenario shown in Figure 4, many different paths are possible for S to reach R. A simple way to benefit from this topology could be to use the two independent paths via nodes A, C, E and via B, D, F. But more complex paths are possible by interleaving transmissions from the lower level of the path to the upper level.¶
PRE may also take advantage of the shared properties of the wireless medium to compensate for the potential loss that is incurred with radio transmissions. For instance, when the source sends to A, B may listen also and get a second chance to receive the frame without an additional transmission. Note that B would not have to listen if it already received that particular frame at an earlier timeslot in a dedicated transmission towards B.¶
The PRE model can be implemented in both centralized and distributed scheduling approaches. In the centralized approach, a Path Computation Element (PCE) scheduler calculates the routes and schedules the communication among the nodes along a circuit such as a Label switched path. In the distributed approach, each node selects its route to the destination, typically using a source routing header. In both cases, at each node in the paths, a default parent and alternative parent(s) should be selected to set up complex tracks.¶
In the following Subsections, all the required operations defined by PRE, namely, Alternative Path Selection, Packet Replication, Packet Elimination and Promiscuous Overhearing, are described.¶
5.3.1. Packet Replication
The objective of PRE is to provide deterministic networking properties: high reliability and bounded latency. To achieve this goal, determinism in every hop of the forwarding paths MUST be guaranteed. By employing a Packet Replication procedure, each node forwards a copy of each data packet to multiple parents: its Default Parent (DP) and multiple Alternative Parents (APs). To do so, each node (i.e., source and intermediate node) transmits the data packet multiple times in unicast to each parent. For instance, in Figure 5, the source node S is transmitting the packet to both parents, nodes A and B, in two different timeslots within the same TSCH slotframe. An example TSCH schedule is shown in Figure 6. Thus, the packet eventually obtains parallel paths to the destination.¶
5.3.2. Packet Elimination
The replication operation increases the traffic load in the network, due to packet duplications. Thus, a Packet Elimination operation SHOULD be applied at each RPL DODAG level to reduce the unnecessary traffic. To this aim, once a node receives the first copy of a data packet, it discards the subsequent copies. Because the first copy that reaches a node is the one that matters, it is the only copy that will be forwarded upward. Then, once a node performs the Packet Elimination operation, it will proceed with the Packet Replication operation to forward the packet toward the RPL DODAG Root.¶
5.3.3. Promiscuous Overhearing
Considering that the wireless medium is broadcast by nature, any neighbor of a transmitter may overhear a transmission. By employing the Promiscuous Overhearing operation, a DP and some AP(s) eventually have more chances to receive the data packets. In Figure 7, when node A is transmitting to its DP (node C), the AP (node D) and its sibling (node B) may decode this data packet as well. As a result, by employing corellated paths, a node may have multiple opportunities to receive a given data packet. This feature not only enhances the end-to-end reliability but also it reduces the end-to-end delay and increases energy efficiency.¶
7. IANA Considerations
This document has no IANA actions.¶
The authors wish to thank¶
10.1. Normative References
- Thubert, P., "An Architecture for IPv6 over the TSCH mode of IEEE 802.15.4", Work in Progress, Internet-Draft, draft-ietf-6tisch-architecture-28, , <https://tools.ietf.org/html/draft-ietf-6tisch-architecture-28>.
- Thubert, P., Cavalcanti, D., Vilajosana, X., and C. Schmitt, "Reliable and Available Wireless Technologies", Work in Progress, Internet-Draft, draft-thubert-raw-technologies-04, , <https://tools.ietf.org/html/draft-thubert-raw-technologies-04>.
- Papadopoulos, G., Thubert, P., Theoleyre, F., and C. Bernardos, "RAW use cases", Work in Progress, Internet-Draft, draft-bernardos-raw-use-cases-03, , <https://tools.ietf.org/html/draft-bernardos-raw-use-cases-03>.
- Grossman, E., Ed., "Deterministic Networking Use Cases", RFC 8578, DOI 10.17487/RFC8578, , <https://www.rfc-editor.org/info/rfc8578>.
- Ratliff, S., Jury, S., Satterwhite, D., Taylor, R., and B. Berry, "Dynamic Link Exchange Protocol (DLEP)", RFC 8175, DOI 10.17487/RFC8175, , <https://www.rfc-editor.org/info/rfc8175>.
- Finn, N. and P. Thubert, "Deterministic Networking Problem Statement", RFC 8557, DOI 10.17487/RFC8557, , <https://www.rfc-editor.org/info/rfc8557>.
- Finn, N., Thubert, P., Varga, B., and J. Farkas, "Deterministic Networking Architecture", RFC 8655, DOI 10.17487/RFC8655, , <https://www.rfc-editor.org/info/rfc8655>.
10.2. Informative References
- Thubert, P., Eckert, T., Brodard, Z., and H. Jiang, "BIER-TE extensions for Packet Replication and Elimination Function (PREF) and OAM", Work in Progress, Internet-Draft, draft-thubert-bier-replication-elimination-03, , <https://tools.ietf.org/html/draft-thubert-bier-replication-elimination-03>.
- Mirsky, G., Chen, M., and D. Black, "Operations, Administration and Maintenance (OAM) for Deterministic Networks (DetNet) with IP Data Plane", Work in Progress, Internet-Draft, draft-mirsky-detnet-ip-oam-02, , <https://tools.ietf.org/html/draft-mirsky-detnet-ip-oam-02>.
- Varga, B., Farkas, J., Berger, L., Malis, A., and S. Bryant, "DetNet Data Plane Framework", Work in Progress, Internet-Draft, draft-ietf-detnet-data-plane-framework-04, , <https://tools.ietf.org/html/draft-ietf-detnet-data-plane-framework-04>.
- Farkas, J., Dudda, T., Shapin, A., and S. Sandberg, "5G - Ultra-Reliable Wireless Technology with Low Latency", Work in Progress, Internet-Draft, draft-farkas-5g-raw-00, , <https://tools.ietf.org/html/draft-farkas-5g-raw-00>.
- IETF, "Mobile Ad hoc Networking", , <https://dataTracker.ietf.org/doc/charter-ietf-manet/>.
- IETF, "Deterministic Networking", , <https://dataTracker.ietf.org/doc/charter-ietf-detnet/>.
- IETF, "Source Packet Routing in Networking", , <https://dataTracker.ietf.org/doc/charter-ietf-spring/>.
- IETF, "Bit Indexed Explicit Replication", , <https://dataTracker.ietf.org/doc/charter-ietf-bier/>.
- IETF, "Bidirectional Forwarding Detection", , <https://dataTracker.ietf.org/doc/charter-ietf-bfd/>.
- IETF, "Common Control and Measurement Plane", , <https://dataTracker.ietf.org/doc/charter-ietf-ccamp/>.