Large-Scale Deterministic Network
draft-qiang-detnet-large-scale-detnet-00
This document is an Internet-Draft (I-D).
Anyone may submit an I-D to the IETF.
This I-D is not endorsed by the IETF and has no formal standing in the
IETF standards process.
The information below is for an old version of the document.
| Document | Type |
This is an older version of an Internet-Draft whose latest revision state is "Expired".
|
|
|---|---|---|---|
| Authors | Li Qiang , Bingyang Liu , Liang Geng | ||
| Last updated | 2018-05-29 | ||
| RFC stream | (None) | ||
| Formats | |||
| Stream | Stream state | (No stream defined) | |
| Consensus boilerplate | Unknown | ||
| RFC Editor Note | (None) | ||
| IESG | IESG state | I-D Exists | |
| Telechat date | (None) | ||
| Responsible AD | (None) | ||
| Send notices to | (None) |
draft-qiang-detnet-large-scale-detnet-00
Network Working Group L. Qiang, Ed.
Internet-Draft B. Liu
Intended status: Informational Huawei
Expires: November 30, 2018 L. Geng
China Mobile
May 29, 2018
Large-Scale Deterministic Network
draft-qiang-detnet-large-scale-detnet-00
Abstract
This document presents a Large-scale Deterministic Network (LDN)
system, which consists of Scalable Deterministic Forwarding (SDF) and
Scalable Resource Reservation (SRR).
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 November 30, 2018.
Copyright Notice
Copyright (c) 2018 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.
Qiang, et al. Expires November 30, 2018 [Page 1]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
Table of Contents
1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2
1.1. Requirements Language . . . . . . . . . . . . . . . . . . 2
1.2. Terminology & Abbreviations . . . . . . . . . . . . . . . 2
2. Scalable Deterministic Forwarding . . . . . . . . . . . . . . 3
2.1. Three Queues . . . . . . . . . . . . . . . . . . . . . . 4
2.2. Cycle Identifier Carrying . . . . . . . . . . . . . . . . 5
3. Scalable Resource Reservation . . . . . . . . . . . . . . . . 5
4. Performance Analysis . . . . . . . . . . . . . . . . . . . . 6
4.1. Queueing Delay . . . . . . . . . . . . . . . . . . . . . 6
4.2. Jitter . . . . . . . . . . . . . . . . . . . . . . . . . 6
5. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6
6. Security Considerations . . . . . . . . . . . . . . . . . . . 6
7. Acknowledgements . . . . . . . . . . . . . . . . . . . . . . 6
8. Normative References . . . . . . . . . . . . . . . . . . . . 6
Authors' Addresses . . . . . . . . . . . . . . . . . . . . . . . 7
1. Introduction
This document presents a Large-scale Deterministic Network (LDN)
system, which consists of Scalable Deterministic Forwarding (SDF) and
Scalable Resource Reservation (SRR). The technologies of SDF and SRR
can be used independently.
As [draft-ietf-detnet-problem-statement] indicates, deterministic
forwarding can only apply on flows with well-defined traffic
characteristics. The traffic characteristics of DetNet flow has been
discussed in [draft-ietf-detnet-architecture], that could be achieved
through shaping at Ingress node or up-front commitment by
application. This document assumes that DetNet flows follow some
specific traffic patterns accordingly.
1.1. Requirements Language
The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this
document are to be interpreted as described in RFC 2119.
1.2. Terminology & Abbreviations
This document uses the terminology defined in
[draft-ietf-detnet-architecture].
TSN: Time Sensitive Network
CQF: Cyclic Queuing and Forwarding
Qiang, et al. Expires November 30, 2018 [Page 2]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
LDN: Large-scale Deterministic Network
SDF: Scalable Deterministic Forwarding
SRR: Scalable Resource Reservation
DSCP: Differentiated Services Code Point
EXP: Experimental
T: the length of a cycle
H: the number of hops
K: the size of aggregated resource reservation window
2. Scalable Deterministic Forwarding
DetNet aims at providing deterministic service over large scale
network. In large scale network, it is difficulty to get precise
time synchronization among numerous and diverse devices. As a
compromise, this document assumes that just clock synchronization is
required among devices. That is different devices maintain the same
clock frequency 1/T, but may at the different time as shown in
Figure 1.
<-----T-----> <-----T----->
| | | | | |
Node A +-----------+-----------+ Node A +-----------+-----------+
| | | | | |
Node B +-----------+-----------+ Node B +-----------+-----------+
(i) time synchronization (ii) clock synchronization
Figure 1: Time Synchronization & Clock Synchronization
IEEE 802.1 CQF is an efficient forwarding mechanism in TSN that
guarantees bounded end-to-end latency. CQF is designed for limited
scale network, the time synchronization is required, and the link
propagation delay is required to be smaller than a cycle length T.
Considering the large scale network deployment, the proposed Scalable
Deterministic Forwarding (SDF) permits clock synchronization and link
propagation delay may exceed T. Besides these two points, CQF and
the asynchronous forwarding are very similar.
Qiang, et al. Expires November 30, 2018 [Page 3]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
Figure 2 compares them through an example. Suppose Node A is the
upstream node of Node B. In CQF, packets sent from Node A at cycle
x, will be received by Node B at the same cycle, then further be sent
to downstream node by Node B at cycle x+1. Due to long link
propagation delay and clock synchronization, Node B will receive
packets from Node A at different cycle denoted by y in the SDF, and
Node B swaps the cycles carried in packets with y+1, then sends out
those packets at cycle y+1. This kind of cycle mapping (e.g., x <-->
y+1) exists between any pair of neighbor nodes, can be studied
through just once forwarding. With this mapping, the receiving node
can easily figure out when the received packets should be send out,
the only requirement is to carry the cycle identifier of sending node
in the packets.
| cycle x | cycle x+1 | | cycle x | cycle x+1 |
Node A +-----------+-----------+ Node A +-----------+-----------+
\ \
\packet \packet
\receiving \receiving
\ \
| V | cycle x+1 | | V | cycle y+1|
Node B +-----------+-----------+ Node B +-----------+-----------+
cycle x \packets cycle y \packets
\sending \sending
\ \
\ \
V V
(i) CQF (ii) SDF
Figure 2: CQF & SDF
2.1. Three Queues
In CQF each port needs to maintain 2 (or 3) queues for each class of
flows, one is used to buffer newly received packets, another one is
used to store the packets that are going to be sent out, one more
queue may be needed to avoid output starvation [scheduled-queues].
While in SDF, at least 3 queues are needed.
As Figure 3 illustrated, a node may receive packets sent at two
different cycles from a single upstream node due to the clock
synchronization. Following the timing slot mapping (i.e., x <-->
y+1), packets that carry cycle identifier x should be send out by
Node B at cycle y+1, and packets that carry cycle identifier x+1
should be send out by Node B at cycle y+2. Therefore, two queues are
needed to store the newly received packets, as well as one queue to
Qiang, et al. Expires November 30, 2018 [Page 4]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
store the sending packets. In order to absorb jitter, more queues
also might be necessary.
| cycle x | cycle x+1 |
Node A +-----------+-----------+
\ \
\ \packet
\ \receiving
| V V | |
Node B +-----------+-----------+
cycle y cycle y+1
Figure 3: Three Queues in SDF
2.2. Cycle Identifier Carrying
As former two sections explained, cycle identifier needs to be
carried in packet, so that an appropriate queue can be selected
accordingly. That means 2 bits is needed in the three queues model
of SDF, in order to identify different cycles between a pair of
neighbor nodes. There are several ways to carry this 2 bits cycle
identifier, for example:
o DSCP of IPv4 Header
o Traffic Class of IPv6 Header
o EXP of MPLS Header
o EtherType of Ethernet Header
o IPv6 Extension Header
o TLV of SRv6
o EXP of MPLS-SR Header
3. Scalable Resource Reservation
SDF must work with some resource reservation mechanisms, that can be
the proposed Scalable Resource Reservation (SRR) or other mechanisms.
Resource reservation guarantees the necessary network resources when
deterministic flows are scheduled. Network nodes have to record how
many network resources are reserved for a specific flow from when it
starts to when it ends (e.g., <flow_identifier, reserved_resource,
start_time, end_time>). Maintaining per-flow resource reservation
status may be acceptable to edge nodes, but un-acceptable to core
Qiang, et al. Expires November 30, 2018 [Page 5]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
nodes. [draft-ietf-detnet-architecture] pointed out that aggregation
must be supported for scalability.
[Details of SRR is TBD.]
4. Performance Analysis
4.1. Queueing Delay
End-to-end queueing delay's expectation is 1.5*T*H, where H is the
number of hops.
[Detailed Analysis is TBD]
4.2. Jitter
Jitter's upper bound is 2*T.
[Detailed Analysis is TBD]
5. IANA Considerations
This document makes no request of IANA.
6. Security Considerations
Security issues have been carefully considered in
[draft-ietf-detnet-security]. More discussion is TBD.
7. Acknowledgements
TBD.
8. Normative References
[draft-ietf-detnet-architecture]
"DetNet Architecture", <https://datatracker.ietf.org/doc/
draft-ietf-detnet-architecture/>.
[draft-ietf-detnet-dp-sol]
"DetNet Data Plane Encapsulation",
<https://datatracker.ietf.org/doc/
draft-ietf-detnet-dp-sol/>.
[draft-ietf-detnet-problem-statement]
"DetNet Problem Statement",
<https://datatracker.ietf.org/doc/
draft-ietf-detnet-problem-statement/>.
Qiang, et al. Expires November 30, 2018 [Page 6]
Internet-Draft Supervised Heterogeneous Network Slicing May 2018
[draft-ietf-detnet-security]
"DetNet Security Considerations",
<https://datatracker.ietf.org/doc/
draft-ietf-detnet-security/>.
[draft-ietf-detnet-use-cases]
"DetNet Use Cases", <https://datatracker.ietf.org/doc/
draft-ietf-detnet-use-cases/>.
[RFC2119] Bradner, S., "Key words for use in RFCs to Indicate
Requirement Levels", BCP 14, RFC 2119,
DOI 10.17487/RFC2119, March 1997,
<https://www.rfc-editor.org/info/rfc2119>.
[scheduled-queues]
"Scheduled queues, UBS, CQF, and Input Gates",
<http://www.ieee802.org/1/files/public/docs2015/
new-nfinn-input-gates-0115-v04.pdf>.
Authors' Addresses
Li Qiang (editor)
Huawei
Beijing
China
Email: qiangli3@huawei.com
Bingyang Liu
Huawei
Beijing
China
Email: liubingyang@huawei.com
Liang Geng
China Mobile
Beijing
China
Email: gengliang@chinamobile.com
Qiang, et al. Expires November 30, 2018 [Page 7]