YANG Data Models for Network Resource Partitions
(NRPs)Huawei Technologies101 Software Avenue, Yuhua DistrictNanjingJiangsu210012Chinalana.wubo@huawei.comHuawei TechnologiesDivyashree Techno ParkBangaloreKarnataka560066Indiadhruv.ietf@gmail.comJuniper Networksvbeeram@juniper.netCisco Systemstsaad.net@gmail.comZTE Corporationpeng.shaofu@zte.com.cn
Routing Area
RFC 9543 describes a framework for Network Slices in networks built
from IETF technologies. In this framework, the network resource
partition (NRP) is introduced as a collection of network resources
allocated from the underlay network to carry a specific set of Network
Slice Service traffic and meet specific Service Level Objective (SLO)
and Service Level Expectation (SLE) characteristics.This document defines two YANG data models for Network Resource
Partitions (NRPs): a network-level model for policy configuration by a
Network Slice Controller, and a device-level model for configuration of
individual network elements. These models enable automated provisioning
of NRPs in IP/MPLS and Segment Routing (SR) networks, supporting
scalable realization of RFC 9543 Network Slice Services. describes a framework for Network Slices in
networks built from IETF technologies. As specified in Section 7.4 , an NRP is a collection of resources identified in
the underlay network to support the RFC 9543 Network Slice Service to
meet the slice Service Level Objectives (SLOs) and Service Level
Expectations (SLEs) characteristics and network scalability. This
document follows the Network Slice Service defined in , and all references to "Network Slice" refer to that
same context.Considering the NRPs realizations in IP/MPLS and Segment Routing (SR)
networks, and describe NRP mechanisms of
control plane, data plane, and management plane to provide specific
forwarding treatment (scheduling, drop policy, resource usage) to the
Slice Service packets associated with an NRP. Specifically, for
instantiation of the device-specific and network wide NRPs, Section 3.5
of introduces the construct of
NRP Policy.Based on these descriptions of NRP, this document defines two YANG
modules:ietf-nrp: A network-level model used by a Network Slice
Controller (NSC) (defined in Section 6.3 )
to manage NRP instances for Network Slice Service realizations. This
is a network configuration model according to the YANG model
classification of .ietf-nrp-device: A device-level model used by a network
controller to set NRP parameters on an individual device, including
device- specific interface configurations. This is a device
configuration model by the classification.The NRPs models conform to the Network Management Datastore
Architecture (NMDA) and are designed to support
both configuration and operational state retrieval.The keywords "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
"SHOULD", "SHOULD NOT", "RECOMMENDED", "NOT RECOMMENDED", "MAY", and
"OPTIONAL" in this document are to be interpreted as described in BCP14,
, when, and only when,
they appear in all capitals, as shown here.The following terms are defined in , , and and are used in this
specification: configuration datastate dataaugmentdata modeldata nodenetwork topologyThe terminology for describing YANG data models is found in . The tree diagram used in this document follows the
notation defined in .Additionally, this document uses the following terms:Network Resource Partition (NRP): A collection of network
resources allocated from the underlay network to support one or more
RFC 9543 Network Slice Services, meeting specific SLO/SLE
requirements. Defined in Section 7.1 of .NRP Policy: A configuration construct that specifies the rules
for creating and maintaining an NRP, including resource allocation,
selector identification, and topology association. NRP Instance: The operational realization of an NRP Policy on
network devices, representing the actual allocated resources and
configured forwarding behaviors.NRP Selector ID: A data plane identifier used by network devices
to classify packets belonging to a specific NRP and apply the
corresponding forwarding treatment.PHB Profile: A named set of Per-Hop Behavior parameters that
define the forwarding treatment (e.g., scheduling, drop policy) for
packets associated with an NRP. Profiles are locally defined by the
network operator.NRP-Enabled Device: A network element that supports NRP data
plane identification and resource partitioning mechanisms.Network Slice Controller (NSC): A functional component
responsible for translating Network Slice Service requests into
network configuration, including NRP management. Defined in Section
6.3 of .As defined in Section 7.1 , the Network
Slice Controller (NSC) can determine which specific connectivity
constructs from one or more slices could be grouped together upon
Slice Service requests. This could be based on a specific set of SLOs
and SLEs, or on any administrative or operational policy. The NSC can
further map these connectivity constructs onto an NRP. It also
constructs and distributes the network wide consistent NRP model to
the relevant controllers, and in turn the controllers distribute the
NRP device model to the NRP-enabled devices in the underlay network.
shows the interfaces to which the two models
are applied.The general operations of NRPs are as follows:NRPs instantiation: Depending on the Slice Service types,
network resources status, and also the operator's policies, there
can be at least two approaches. One method is to create an NRP
instance before the network controller processes a Network Slice
Service request. Another one is that the network controller may
decide creating an NRP instance while implementing the Network
Slice Service request.NRPs modification: When the capacity of an existing NPR link is
close to maximum capacity, the bandwidth of the link could be
increased. And when an NRP links or nodes resources are
insufficient, new NRP links and nodes could be added.NRPs Deletion: If an NSC determines that no Slice Service is
using an NRP, the NSC can delete the NRP instance. Another example
is when it is necessary to merge NRPs.NRPs Monitoring: The NSC can use the NRPs model to track and
monitor NRPs resource status and usage.An NRP Policy is a configuration construct that specifies the rules
for creating and maintaining a Network Resource Partition. It defines
how network resources are allocated, how traffic is identified, which
topology is used, and what forwarding treatment is applied. These
rules apply to both device-level and network-level configurations.
An NRP Policy includes the following key components:Topology customization rules: Determine the set of nodes,
links, and network functions that belong to the NRP.Data plane rules: Specify the NRP selector ID, QoS profiles,
and bandwidth reservations for traffic identification and
forwarding treatment.Control plane rules: Define resource reservation, resource
sharing, reservation preferences policy, etc. NRP Policies support three resource partition modes:Data Plane Partition: Physical network resources (e.g.,
bandwidth) are partitioned on network devices. Devices enable NRP
data plane mechanisms to apply Per-Hop Behavior (PHB) to
packets.Control Plane Partition: Resource reservation state for each
NRP is maintained in NRP-aware TE devices or network controllers,
without physical resource isolation.Hybrid Partition: Combines both data plane and control plane
partitioning for full resource isolation and stateful
management.All modes require NRP topology specification. Modes (a) and (c)
require NRP selector identification for packet classification. Modes
(b) and (c) require distributed or centralized resource reservation
management. When IGP routing is used with mode (c), NRP-aware IGP
extensions may be employed for scalability.For further implementation considerations, the concepts align with
the NRP mechanisms described in and .The high-level model structure of NRP policy as modeled in this
document is shown in :The "networks" container from the "ietf-network" module provides a placeholder for an inventory of nodes in
the network. This container is augmented to include a set of NRP
policies.The "policy-profiles" container provides a list of PHB policy
profile entries. Each of these entries can be referenced by one or
more NRPs. A "phb-profile-identifier" entry can have a reference to a
standard PHB profile available on the device or the network
controller.The "nrp-policies" container includes a list of NRP policies. Each
"nrp-policy" entry is identified by a name and holds the set of
attributes needed to instantiate an NRP.The description of the "nrp-policies" data nodes are as follows,
and the other key elements of each nrp-policy entry are discussed in
the following sub-sections."nrp-id": Is an identifier that is used to uniquely identify an
NRP instance within an NSC network scope, which is created by the
enforcement of the "nrp-policy"."mode": Refers to control plane partition, data plane
partition, or hybrid partition.The "resource-reservation" container specifies bidirectional
bandwidth allocation for an NRP, expressed as either an absolute
value or a percentage of link capacity. While currently limited to
bandwidth, the container is designed to accommodate future
extensions such as resource sharing or protection mechanisms
described in .The bandwidth reservation specified at the policy level serves as
a default for all links in the NRP topology. This can be overridden
by link-specific configuration in the "nrp-topology" container .The NRP selector ID defines data plane encapsulation types and
values used to identify NRP-specific network resources. The
configuration can be overridden by link-specific "selector-id"
configuration in "nrp-topology" ().Implementations MUST support at least one selector type in the
NRP model of data plane partition or hybrid partition. Multiple
selector types MAY be configured to support heterogeneous network
environments. Selector ID values MUST be unique within the network
scope where the NRP operates.The following selector types are supported:IPv4-derived: Uses destination IPv4 prefixes to identify NRP
traffic. Suitable for networks where NRPs are aligned with IP
addressing.IPv6: Provides three sub-options: - Dedicated: Uses a value in the IPv6 Hop-by-Hop extension
header as specified in - SRv6-derived: Uses SRv6 SIDs where the SID structure
encodes the NRP identifier, per . - IPv6-destination-derived: Uses destination IPv6 prefixes,
similar to IPv4-derived.MPLS: Supports two encoding mechanisms: - In-Stack MPLS Network Action (MNA): Embeds the selector in
the MPLS label stack using the format defined in . Supports 13-bit
(NRPS13) or 20-bit (NRPS20/ENRPS20) encodings. - Post-Stack (PSD): Uses a dedicated post-stack label per
.ACL-based: References standard ACLs to match NRP traffic.
Provides flexibility for complex classification rules but may
have higher processing overhead.Per-Hop Behavior (PHB) defines the forwarding treatment for
packets belonging to an NRP, including bandwidth control, congestion
control, and scheduling priority (e.g., Section 3.4 ). The PHB profile and NRP selector are paired
mechanisms: the selector classifies packets, and the referenced PHB
profile applies the corresponding forwarding behavior. Profile
contents are implementation-specific and may include standard IETF
PHBs (e.g., "Assured Forwarding (AF)", "Expedited Forwarding (EF)")
or customized local policies (e.g., "High", "Low", "Standard").The phb-profile configured at the NRP policy level serves as the
default for the entire NRP topology. This may be overridden by
link-specific PHB configuration in the topology container ."nrp-topology" defines a customized NRP topology used for an
NRP.When an NRP supports IGP routing, the topology of the NRP must be
congruent with an IGP instance. The topology used for IGP route
computation and forwarding can be derived using Multi-Topology
Routing (MTR) , ,
and or Flex-algo . shows an example of NRP-1 enabling
"igp-congruent", which indicates that this NRP instance uses the
same IGP topology with the specified "multi-topology-id" or
"algo-id". NRP-1 has different link resource attributes from those
of the IGP, but shares the same nodes and termination points (TPs)
of the IGP topology.The "selection" container consists of a list of select subset of
links of an underlay topology or a pre-built topology.The "filter" container consists of a list of filters where each
entry references a topology filter . The topological
elements that satisfy the membership criteria may override the
default "resource-reservation" and "selector-id" specific nodes. /nw:networks/network/network-id
| +--rw link* [link-ref]
| | ...
| +--rw resource-reservation
| | ...
| +--rw link-partition-type? identityref
| +--rw phb-profile? leafref
+--:(filter)
+--rw filters
+--rw filter* [filter-ref]
+--rw filter-ref
| topo-filter-ref
+--rw resource-reservation
| ...
+--rw selector-id
| ...
+--rw phb-profile? leafref]]>The NRP model can be used to monitor the operational status and
resource usage of NRPs.The device-specific NRPs model is defined in module
"ietf-nrp-device" as shown in .The NRP device YANG data model is only applicable to device
configuration and includes attributes such as QoS policies, resource
reservations, and NRP selector IDs. Specifically, it adds
interface-specific attributes for cases where the QoS policies, NRP
resources, and NRP selector IDs of an interface differ from the global
NRP attributes of the device. shows the tree diagram of the device
NRPs YANG model defined in modules "ietf-nrp-device.yang".The "ietf-nrp" module uses types defined in ,
,, , , , and . file "ietf-nrp@2026-01-22.yang"
module ietf-nrp {
yang-version 1.1;
namespace "urn:ietf:params:xml:ns:yang:ietf-nrp";
prefix nrp;
import ietf-network {
prefix nw;
reference
"RFC 8345: A YANG Data Model for Network Topologies,
Section 6.1";
}
import ietf-network-topology {
prefix nt;
reference
"RFC 8345: A YANG Data Model for Network Topologies,
Section 6.2";
}
import ietf-routing-types {
prefix rt-types;
reference
"RFC 8294: Common YANG Data Types for the Routing Area";
}
import ietf-vpn-common {
prefix vpn-common;
reference
"RFC 9181: A Common YANG Data Model for Layer 2 and Layer 3
VPNs";
}
import ietf-te-packet-types {
prefix te-packet-types;
reference
"RFC 8776: Traffic Engineering Common YANG Types";
}
import ietf-inet-types {
prefix inet;
reference
"RFC 9911: Common YANG Data Types";
}
import ietf-access-control-list {
prefix acl;
reference
"RFC 8519: YANG Data Model for Network Access Control Lists
(ACLs)";
}
import ietf-srv6-types {
prefix srv6-types;
reference
"draft-ietf-spring-srv6-yang: YANG Data Model for SRv6 Base
and Static";
}
import ietf-topology-filter {
prefix topo-filt;
reference
"draft-ietf-teas-yang-topology-filter: YANG Data Model
for Topology Filter";
}
organization
"IETF Traffic Engineering Architecture and Signaling (TEAS)
Working Group";
contact
"WG Web:
WG List:
Editor: Bo Wu
Editor: Dhruv Dhody
Editor: Vishnu Pavan Beeram
Editor: Tarek Saad
Editor: Shaofu Peng
";
description
"This YANG module defines a data model for
Network Resource Partitions (NRPs) management.
Copyright (c) 2026 IETF Trust and the persons identified as
authors of the code. All rights reserved.
Redistribution and use in source and binary forms, with or
without modification, is permitted pursuant to, and subject
to the license terms contained in, the Revised BSD License
set forth in Section 4.c of the IETF Trust's Legal Provisions
Relating to IETF Documents
(https://trustee.ietf.org/license-info).
This version of this YANG module is part of RFC XXXX
(https://www.rfc-editor.org/info/rfcXXXX); see the RFC itself
for full legal notices.";
revision 2026-01-22 {
description
"Initial revision.";
reference
"RFC XXXX: YANG Data Models for Network Resource
Partitions (NRPs)";
}
/*
* I D E N T I T I E S
*/
identity nrp-partition-mode {
description
"Base identity for NRP partition type.";
}
identity control-plane-partition {
base nrp-partition-mode;
description
"NRP control plane partition.";
}
identity data-plane-partition {
base nrp-partition-mode;
description
"NRP data plane partition.";
}
identity hybrid-plane-partition {
base nrp-partition-mode;
description
"Both control and data planes partitions of NRP.";
}
identity nrp-link-partition-type {
description
"Base identity for NRP interface partition type.";
}
identity virtual-interface {
base nrp-link-partition-type;
description
"Identity for NRP virtual interface or sub-interface partition,
e.g., FlexE.";
}
identity queue-partition {
base nrp-link-partition-type;
description
"Identity for NRP queue partition type.";
}
/*
* T Y P E D E F S
*/
typedef acl-ref {
type leafref {
path "/acl:acls/acl:acl/acl:name";
}
description
"Used to reference an ACL.";
}
typedef topo-filter-ref {
type leafref {
path "/nw:networks/topo-filt:topology-filters/"
+ "topo-filt:topology-filter/topo-filt:name";
}
description
"This type is used to reference a Topology Filter.";
reference
"draft-ietf-teas-yang-topology-filter: YANG Data Model
for Topology Filter";
}
/*
* Grouping - NRP Resource Reservation
*/
grouping resource-reservation {
description
"Grouping for NRP resource reservation.";
container resource-reservation {
description
"Container for NRP resource reservation.";
choice max-bw-type {
description
"Choice of maximum bandwidth specification.";
case bw-value {
leaf maximum-bandwidth {
type uint64;
units "bits/second";
description
"The maximum bandwidth allocated to an NRP
- specified as absolute value.";
}
}
case bw-percentage {
leaf maximum-bandwidth-percent {
type rt-types:percentage;
description
"The maximum bandwidth allocated to an NRP
- specified as percentage of link
capacity.";
}
}
}
}
}
/*
* Grouping - NRP Selector Identifier Configuration
*/
grouping selector-id {
description
"Grouping for NRP selector identifier (NRP Selector ID)
configuration.";
container selector-id {
description
"Container for NRP selector ID.";
container ipv4-derived {
description
"Container for IPv4 NRP selector ID.";
leaf-list destination-prefix {
type inet:ipv4-prefix;
description
"Any prefix from the specified set of IPv4
destination prefixes can be the selector ID.";
}
}
container ipv6 {
description
"Container for IPv6 NRP selector ID.";
choice selector-type {
description
"Choices for IPv6 selector ID type.";
case dedicated {
leaf ipv6-hbh-eh {
type uint32;
description
"The selector ID carried in Hop-by-Hop option of
IPv6 extension header.";
reference
"draft-ietf-6man-enhanced-vpn-vtn-id: Carrying
Network Resource (NRP) related Information in IPv6
Extension Header";
}
}
case srv6-derived {
leaf-list srv6-sid {
type srv6-types:srv6-sid;
description
"Any SID from the specified set of SRv6 SID can
be the NRP selector ID.";
reference
"draft-ietf-spring-sr-for-enhanced-vpn: Segment
Routing based Network Resource Partition (NRP) for
Enhanced VPN";
}
}
case ipv6-destination-derived {
leaf-list destination-prefix {
type inet:ipv6-prefix;
description
"Any prefix from the specified set of IPv6
destination prefixes can be the NRP selector ID.";
}
}
}
}
container mpls {
description
"Container for MPLS NRP selector ID. Supports both
In-Stack (MNA) and Post-Stack (PSD) encoding mechanisms.";
choice mpls-encapsulation {
description
"Selects the MPLS NRP Selector encoding mechanism.";
case mna-in-stack {
description
"In-Stack NRP Selector action.
Refers to draft-ietf-mpls-mna-nrp-selector.";
leaf in-stack-identifier {
type uint32;
must
"(../identifier-format='nrps13' and . <= 8191) or
(../identifier-format='nrps20' and . <= 1048575) or
(../identifier-format='enrps20' and . <= 1048575)" {
error-message
"NRP Selector value exceeds format range";
}
description
"The NRP Selector value. The valid range depends on the
encoding format (13-bit or 20-bit) selected.";
}
leaf identifier-format {
type enumeration {
enum nrps13 {
description
"13-bit NRP Selector (NRPS13).
Value range: 0-8191.";
}
enum nrps20 {
description
"20-bit NRP Selector (NRPS20).
Value range: 0-1048575.";
}
enum enrps20 {
description
"20-bit Extended NRP Selector (ENRPS20) compatible
with Entropy Label. Value range: 0-1048575.";
}
}
description
"Specifies the encoding format for the In-Stack
NRP Selector.";
}
}
case post-stack {
description
"Post-Stack NRP Selector (PS-NRP).
Refers to draft-ietf-mpls-mna-psd-nrp-selector.";
leaf post-stack-identifier {
type uint32 {
range "0..4294967295"; // 2^32 - 1
}
description
"The 32-bit Post-Stack NRP Selector.";
}
}
}
}
leaf-list acl-ref {
type acl-ref;
description
"NRP selection is done based on the specified list of
ACLs.";
reference
"RFC 8519: YANG Data Model for Network Access Control Lists
(ACLs)";
}
}
}
/*
* Grouping - NRP QoS Per-Hop Behavior (PHB) profiles
*/
grouping qos-profiles {
description
"Grouping for NRP QoS profiles.";
container qos-profiles {
description
"Container for profiles.";
list phb-profile-identifier {
key "id";
description
"List of PHB profiles.";
leaf id {
type string;
description
"Uniquely identifies the PHB.
The profile only has significance within the service
provider's administrative domain";
}
}
}
}
/*
* Grouping - NRP QoS Per-Hop Behavior (PHB) profile
*/
grouping qos-phb-profile {
description
"Grouping for NRP QoS Per-Hop Behavior (PHB) profile.";
leaf phb-profile {
type leafref {
path "/nw:networks/nrp:nrp-policies/"
+ "nrp:qos-profiles/nrp:phb-profile-identifier/nrp:id";
}
description
"PHB profile identifier, specifying the forwarding treatment
of packets belonging to a specific NRP selector ID, such as
bandwidth control, congestion control
(e.g., Section 3.4 of [RFC3644]).
The PHB may be standard PHB, such as Assured Forwarding (AF),
Expedited Forwarding (EF), or a customized local policy,
such as 'High', 'Low', 'Standard'.";
}
}
/*
* Grouping - NRP IGP congruent
*/
grouping igp-congruent {
description
"Grouping for NRP IGP congruent attributes.";
container igp-congruent {
presence "Indicates NRP IGP congruency.";
description
"The presence of the container node describes NRP IGP
congruent, which indicates that the NRP instance uses the
same IGP topology with the specified 'multi-topology-id'
and 'algo-id'. That is, the nodes and termination point of
the NRP topology and the IGP topology are the same, while
the link attributes of the NRP are different from those of
the IGP.";
leaf multi-topology-id {
type uint32;
description
"Indicates the MT-id of the NRP IGP instance.";
reference
"RFC 5120: M-ISIS: Multi Topology (MT) Routing in
Intermediate System to Intermediate Systems (IS-ISs)
RFC 4915: Multi-Topology (MT) Routing in OSPF";
}
leaf algo-id {
type uint32;
description
"Indicates the algo-id of the NRP IGP instance.";
reference
"RFC 9350: IGP Flexible Algorithm";
}
leaf sharing {
type boolean;
description
"'true' if the NRP IGP instance can be shared with
other NRPs;
'false' if the NRP IGP instance is dedicated
to this NRP.";
}
}
}
/*
* Grouping - NRP Topology Filter
*/
grouping topology-filter {
description
"Grouping for NRP filter topology.";
container filters {
description
"Container for filters.";
list filter {
key "filter-ref";
description
"List of filters.";
leaf filter-ref {
type topo-filter-ref;
description
"Reference to a specific topology filter from the
list of global topology filters.";
}
uses resource-reservation;
uses selector-id;
uses qos-phb-profile;
}
}
}
/*
* Grouping - NRP Select Topology
*/
grouping select-topology {
description
"NRP topology specified by selection.";
container select {
description
"The container of NRP select topology.";
list topology-group {
key "group-id";
description
"List of groups for NRP topology elements (node or links)
that share common attributes.";
leaf group-id {
type string;
description
"The NRP topology group identifier.";
}
/*The base topology reference. All links in
the list below must belong to this specific
network.*/
uses nw:network-ref;
list link {
key "link-ref";
description
"A list of links with common attributes that are
part of the referenced network";
leaf link-ref {
type leafref {
path "/nw:networks/nw:network"
+ "[nw:network-id=current()/../../nrp:network-ref]"
+ "/nt:link/nt:link-id";
}
description
"A reference to a link in the base topology specified
by the 'network-ref' sibling node.";
}
}
uses resource-reservation;
leaf link-partition-type {
type identityref {
base nrp-link-partition-type;
}
description
"Indicates the resource reservation type of an NRP link.";
}
uses qos-phb-profile;
}
}
}
/*
* Grouping - NRP Policy
*/
grouping nrp-pol {
description
"Grouping for NRP policies.";
container nrp-policies {
description
"Container for nrp policies.";
uses qos-profiles;
list nrp-policy {
key "name";
unique "nrp-id";
description
"List of NRP policies.";
leaf name {
type string;
description
"A string that uniquely identifies the NRP policy.";
}
leaf nrp-id {
type uint32;
description
"A 32-bit ID that uniquely identifies the NRP
created by the enforcement of this NRP policy.";
}
leaf mode {
type identityref {
base nrp-partition-mode;
}
description
"Indicates the resource partition mode of the NRP, such
as control plane partition, data plane partition,
or hybrid partition.";
}
uses resource-reservation;
uses selector-id;
uses qos-phb-profile;
container topology {
description
"Container for NRP topology.";
uses igp-congruent;
choice topology-type {
description
"Choice of NRP topology type.";
case selection {
uses select-topology;
}
case filter {
uses topology-filter;
}
}
}
}
}
}
/*
* Grouping - NRP Selector ID State
*/
grouping selector-id-state {
description
"The grouping of NRP selector ID state.";
container selector-id {
config false;
description
"The container of NRP selector ID.";
leaf srv6 {
type srv6-types:srv6-sid;
description
"SRv6 SID value.";
}
leaf mpls-nrp-selector {
type uint32;
description
"MPLS NRP selector value.";
}
}
}
// nrp-link-statistics
grouping statistics-per-link {
description
"Statistics attributes per NRP link.";
container statistics {
config false;
description
"Statistics for NRP link.";
uses vpn-common:service-status;
uses nrp-bandwidth-metrics;
uses te-packet-types:one-way-performance-metrics-packet;
}
}
/*
* Grouping - NRP node attributes
*/
grouping node-nrp-attributes {
description
"NRP node scope attributes.";
container nrp-node-attributes {
config false;
description
"Containing NRP attributes.";
uses selector-id-state;
}
}
/*
* Grouping - NRP Link Attributes
*/
grouping link-nrp-attributes {
description
"NRP link scope attributes.";
container nrp-link-attributes {
config false;
description
"Contains NRP link attributes.";
leaf link-partition-type {
type identityref {
base nrp-link-partition-type;
}
description
"Indicates the resource partition type of an NRP link.";
}
leaf bandwidth-value {
type uint64;
units "bits/second";
description
"Bandwidth allocation for the NRP as absolute value.";
}
uses selector-id-state;
uses statistics-per-link;
}
}
/*
* Grouping - NRP Bandwidth Metrics
*/
grouping nrp-bandwidth-metrics {
description
"Grouping for NRP bandwidth metrics.";
leaf one-way-available-bandwidth {
type uint64;
units "bits/second";
description
"Available bandwidth that is defined to be NRP link
bandwidth minus bandwidth utilization.";
}
leaf one-way-utilized-bandwidth {
type uint64;
units "bits/second";
description
"Bandwidth utilization that represents the actual
utilization of the link (i.e., as measured in the router).";
}
}
/*
* Grouping - NRPs Node Attributes
*/
grouping node-nrp-list-attributes {
description
"Grouping for NRPs nodes attributes.";
container node-nrp-list {
config false;
description
"Containing NRPs attributes.";
list nrp {
key "nrp-id";
description
"List of NRPs.";
leaf nrp-id {
type uint32;
description
"NRP identifier";
}
uses node-nrp-attributes;
}
}
}
/*
* Grouping - NRPs Link Attributes
*/
grouping link-nrp-list-attributes {
description
"Grouping for NRPs link attributes.";
container link-nrp-list {
config false;
description
"Contains NRPs link attributes.";
list nrp {
key "nrp-id";
description
"List of NRPs.";
leaf nrp-id {
type uint32;
description
"NRP identifier";
}
uses link-nrp-attributes;
}
}
}
// nrp-network-type
grouping nrp-network-type {
description
"Identifies the network type to be NRP.";
container nrp {
presence "Indicates NRP network topology.";
description
"The presence of the container node indicates NRP network.";
}
}
/*
* Augment - Network Resource Partition Policies.
*/
augment "/nw:networks" {
description
"Augments networks with NRP policies.";
uses nrp-pol;
}
/*
* Augment - NRP type.
*/
augment "/nw:networks/nw:network/nw:network-types" {
description
"Indicates the network type of NRP";
uses nrp-network-type;
}
/*
* Augment - NRP node operational status.
*/
augment "/nw:networks/nw:network/nw:node" {
description
"Augments node with NRP state attributes.";
choice nrp-attachment-type {
description
"Determines how NRP information is attached to node:
single-nrp for NRP-specific topology, multi-nrp for
native topology with multiple NRPs.";
case single-nrp {
when '../nw:network-types/nrp:nrp' {
description
"Condition: only applies to NRP network topologies.";
}
uses node-nrp-attributes;
}
case multi-nrp {
description
"Native topology with NRPs node operational status.";
uses node-nrp-list-attributes;
}
}
}
/*
* Augment - NRP link operational status.
*/
augment "/nw:networks/nw:network/nt:link" {
description
"Augments link with NRP state attributes.";
choice nrp-attachment-type {
description
"Determines how NRP information is attached to link:
single-nrp for NRP-specific topology, multi-nrp for
native topology with multiple NRPs.";
case single-nrp {
when '../nw:network-types/nrp:nrp' {
description
"Condition: only applies to NRP network topology.";
}
uses link-nrp-attributes;
}
case multi-nrp {
/*Native topology with NRPs link operational status.*/
description
"Native topology with NRPs link operational status.";
uses link-nrp-list-attributes;
}
}
}
}
]]>The NRP YANG module for devices ("ietf-nrp-device") reuses the qos
policy, "resource-reservation", "selector-id" grouping defined in ,and adds interface-specific NRP attributes.The device NRPs YANG module also imports the following module(s):
ietf-interfaces defined in . file "ietf-nrp-device@2025-03-02.yang"
module ietf-nrp-device {
yang-version 1.1;
namespace "urn:ietf:params:xml:ns:yang:ietf-nrp-device";
prefix nrp-dev;
/* Import IETF interface module */
import ietf-interfaces {
prefix if;
reference
"RFC8343: A YANG Data Model for Interface Management";
}
/* Import NRPs module */
import ietf-nrp {
prefix nrp;
reference
"RFCXXXX: YANG Data Models for Network Resource
Partitions (NRPs)";
}
organization
"IETF Traffic Engineering Architecture and Signaling (TEAS)
Working Group";
contact
"WG Web:
WG List:
Editor: Bo Wu
Editor: Dhruv Dhody
Editor: Vishnu Pavan Beeram
Editor: Tarek Saad
Editor: Shaofu Peng
";
description
"This YANG module defines a data model for Network Resource
Partitions (NRPs) device configurations and states. The model
fully conforms to the Network Management Datastore
Architecture (NMDA).
Copyright (c) 2025 IETF Trust and the persons identified as
authors of the code. All rights reserved.
Redistribution and use in source and binary forms, with or
without modification, is permitted pursuant to, and subject to
the license terms contained in, the Revised BSD License set
forth in Section 4.c of the IETF Trust's Legal Provisions
Relating to IETF Documents
(https://trustee.ietf.org/license-info).
This version of this YANG module is part of RFC XXXX
(https://www.rfc-editor.org/info/rfcXXXX); see the RFC itself
for full legal notices.";
// RFC Ed.: replace XXXX with actual RFC number and
// remove this note.
// RFC Ed.: update the date below with the date of RFC
// publication and remove this note.
revision 2025-03-02 {
description
"Initial revision.";
reference
"RFCXXXX: YANG Data Models for Network Resource
Partitions (NRPs)";
}
/*
* Grouping - NRP QoS Per-Hop Behavior (PHB) profile
*/
grouping qos-phb-profile {
description
"Grouping for NRP QoS Per-Hop Behavior (PHB) profile.";
leaf phb-profile {
type leafref {
path
"/nrp-policies/qos-profiles/phb-profile-identifier/id";
}
description
"PHB profile identifier, specifying the forwarding treatment
of packets belonging to a specific NRP selector identifier,
such as bandwidth control, congestion control
(e.g., Section 3.4 of [RFC3644]).
The PHB may be standard PHB, such as Assured
Forwarding (AF), Expedited Forwarding (EF), or
a customized local policy, such as 'High', 'Low',
'Standard'.";
}
}
/* NRP device configuraiton */
container nrp-policies {
description
"Container for nrp policies.";
uses nrp:qos-profiles;
list nrp-policy {
key "name";
unique "nrp-id";
description
"List of NRP policies.";
leaf name {
type string;
description
"A string that uniquely identifies the NRP policy.";
}
leaf nrp-id {
type uint32;
description
"A 32-bit ID that uniquely identifies the NRP
created by the enforcement of this NRP policy.";
}
uses nrp:resource-reservation;
uses nrp:selector-id;
uses qos-phb-profile;
uses nrp:igp-congruent;
/* NRP Interface Configuration Data */
container interfaces {
description
"NRP interfaces global configuration.";
list interface {
key "interface";
description
"The list of interfaces enabled for NRP.";
leaf interface {
type if:interface-ref;
description
"NRP interface name.";
}
uses nrp:resource-reservation;
uses nrp:selector-id;
uses qos-phb-profile;
}
}
}
}
}
]]> analyzes the
scalability considerations of the control plane and data plane in the
NRPs deployment. This section complements some scalability
considerations with the model and the possible implications on
deployment or implementation.As discussed in Section 2 of , the number of Network Resource
Partitions (NRPs) required depends on the deployment scenario:
multi-service networks typically need around ten NRPs, whereas
industrial-vertical and cloud-network expansions can require hundreds to
thousands of NRPs to support correspondingly large numbers of Network
Slices. Compared to a single NRP, all these numbers of NRPs pose
challenges for management and operations, including NRP planning,
provisioning, monitoring, and Network Slice Service provisioning and
assurance based on NRPs. Section 3 of describes the detailed realization
process of multiple NRPs.For better scalability, the NRP YANG model design considers the
following aspects:Planning: A reusable "qos-profile" groups Service-Level
Objectives (SLOs) once and applies them to any NRP, guaranteeing
consistent QoS policy across the domain while eliminating per-NRP
duplication. "filter/selection" enables multiple NRPs to share a
common topology, confining per-NRP state to the differences rather
than replicating the entire topology. Network operators should
create new NRPs only when strict resource isolation is required,
ensuring that network-wide resources remain efficiently
utilized.Provisioning: The network controller, not the Network Slice
Controller (NSC), is responsible for the actual resource allocation
of NRPs on devices. It assigns NRPs to the relevant interfaces,
reserve the bandwidth, and set data-plane selector IDs. NSC conveys
the intent through the NRP network model, specifically via the
"nrp-policy", and the network controller then realize the requested
allocation. And for data-plane selector IDs, Section 5.2 of recommends selecting a
data-plane mechanism that minimizes per-NRP configuration. A
dedicated IPv6 selector identifier is one example that achieves this
objective.Monitoring: The model supports two complementary monitoring
approaches: Per-NRP view: Operational state scoped to a single "nrp-id",
letting the controller present the logical network dedicated to
that NRP.Network-wide view: An extension of the RFC 8345 topology
model that exposes the mapping between the underlay network and
all associated NRPs, including per-node and per-link NRP
status.The YANG models defined in this document are designed to be accessed
via network management protocols such as NETCONF or RESTCONF . The lowest
NETCONF layer is the secure transport layer, and the
mandatory-to-implement secure transport is Secure Shell (SSH) . The lowest RESTCONF layer is HTTPS, and the
mandatory-to-implement secure transport is TLS .The NETCONF access control model provides
the means to restrict access for particular NETCONF or RESTCONF users to
a preconfigured subset of all available NETCONF or RESTCONF protocol
operations and content.There are a number of data nodes defined in this YANG model that are
writable/creatable/deletable (i.e., config true, which is the default).
These data nodes may be considered sensitive or vulnerable in some
network environments. Write operations (e.g., edit-config) to these data
nodes without proper protection can have a negative effect on network
operations.nrp-policies: Unauthorized creation or modification of NRP policies
can allocate excessive network resources (bandwidth, buffers, forwarding
entries), causing denial of service to other services. Implementations
MUST enforce resource limits and validate quota constraints before
applying NRP configurations. NETCONF access control SHOULD restrict NRP
policy creation to authorized network-admin roles only.selector-id and phb-profile: Misconfiguration of NRP selector IDs or
PHB profiles may cause traffic to be forwarded on the wrong NRP,
violating isolation guarantees and SLOs. Implementations MUST validate
selector ID uniqueness within the network scope and prevent overlapping
PHB configurations that could create ambiguous forwarding rules.topology: Modification of NRP topology (links, nodes, filters) could
reroute traffic through unintended paths, bypassing security policies or
causing congestion. Operators MUST apply access controls to topology
modifications and audit all changes. The model's operational state
SHOULD be monitored to detect unauthorized topology alterations.The entries in the nodes above include the whole network
configurations corresponding with the NRP, and indirectly create or
modify the PE or P device configurations. Unexpected changes to these
entries could lead to service disruption and/or network misbehavior.Sensitive Data Exposure: Operational state data (e.g., utilized
bandwidth, delay statistics) may reveal sensitive information about
network load and customer traffic patterns. Access to operational state
SHOULD be restricted using NACM rules, limiting visibility to authorized
monitoring systems.This document registers a URI in the IETF XML registry . Following the format in ,
the following registration is requested to be made:This document requests to register a YANG module in the YANG Module
Names registry .The authors would like to thank Krzysztof Szarkowicz, Jie Dong, Qin
Wu, Yao Zhao, Zhenbing Li, Adrian Farrel, Tom Petch, Xuesong Geng, Italo
Busi,and many others for their helpful comments and suggestions.The following individuals, authors of and , contributed to this consolidated
document: All issues identified during Working Group review have been resolved.
This section is intentionally left blank for the final publication.This section contains an example of an instance data tree in JSON
encoding . | | 2-1-1 | |
| | 1-2-1 | |<----------------| | 2-0-1 | |
\-/ 1-3-1 \-/ \-/ 2-3-1 \-/
| /----\ | | /----\ |
+---| |---+ +---| |---+
\----/ \----/
| | | |
| | | |
| | | |
| | +------------+ | |
| | | D3 | | |
| | /-\ /-\ | |
| +----->| | 3-1-1 | |-------+ |
+---------| | 3-2-1 | |<---------+
\-/ \-/
| |
+------------+
]]>The following example configures a hybrid NRP (NRP1) with SRv6
selector and Flex-Algo based topology.Another example of an NRP that supports the control plane partition
mode is shown in the following figure.The following example configures a router with two NRP instances
using IPv6 Hop-by-Hop extension header options for traffic
identification. NRP-100 uses selector ID 100 for high-priority service
traffic, while NRP-200 uses selector ID 200 for standard services. shows the full tree diagram of the NRPs
YANG model defined in module "ietf-nrp.yang". /nw:networks/network/network-id
| +--rw link* [link-ref]
| | +--rw link-ref leafref
| +--rw resource-reservation
| | +--rw (max-bw-type)?
| | +--:(bw-value)
| | | +--rw maximum-bandwidth?
| | | uint64
| | +--:(bw-percentage)
| | +--rw maximum-bandwidth-percent?
| | rt-types:percentage
| +--rw link-partition-type? identityref
| +--rw phb-profile? leafref
+--:(filter)
+--rw filters
+--rw filter* [filter-ref]
+--rw filter-ref
| topo-filter-ref
+--rw resource-reservation
| +--rw (max-bw-type)?
| +--:(bw-value)
| | +--rw maximum-bandwidth?
| | uint64
| +--:(bw-percentage)
| +--rw maximum-bandwidth-percent?
| rt-types:percentage
+--rw selector-id
| +--rw ipv4-derived
| | +--rw destination-prefix*
| | inet:ipv4-prefix
| +--rw ipv6
| | +--rw (selector-type)?
| | +--:(dedicated)
| | | +--rw ipv6-hbh-eh?
| | | uint32
| | +--:(srv6-derived)
| | | +--rw srv6-sid*
| | | srv6-types:srv6-sid
| | +--:(ipv6-destination-derived)
| | +--rw destination-prefix*
| | inet:ipv6-prefix
| +--rw mpls
| | +--rw (mpls-encapsulation)?
| | +--:(mna-in-stack)
| | | +--rw in-stack-identifier?
| | | | uint32
| | | +--rw identifier-format?
| | | enumeration
| | +--:(post-stack)
| | +--rw post-stack-identifier?
| | uint32
| +--rw acl-ref* acl-ref
+--rw phb-profile? leafref
augment /nw:networks/nw:network/nw:network-types:
+--rw nrp!
augment /nw:networks/nw:network/nw:node:
+--rw (nrp-attachment-type)?
+--:(single-nrp)
| +--ro nrp-node-attributes
| +--ro selector-id
| +--ro srv6? srv6-types:srv6-sid
| +--ro mpls-nrp-selector? uint32
+--:(multi-nrp)
+--ro node-nrp-list
+--ro nrp* [nrp-id]
+--ro nrp-id uint32
+--ro nrp-node-attributes
+--ro selector-id
+--ro srv6? srv6-types:srv6-sid
+--ro mpls-nrp-selector? uint32
augment /nw:networks/nw:network/nt:link:
+--rw (nrp-attachment-type)?
+--:(single-nrp)
| +--ro nrp-link-attributes
| +--ro link-partition-type? identityref
| +--ro bandwidth-value? uint64
| +--ro selector-id
| | +--ro srv6? srv6-types:srv6-sid
| | +--ro mpls-nrp-selector? uint32
| +--ro statistics
| +--ro status
| | +--ro admin-status
| | | +--ro status? identityref
| | | +--ro last-change? yang:date-and-time
| | +--ro oper-status
| | +--ro status? identityref
| | +--ro last-change? yang:date-and-time
| +--ro one-way-available-bandwidth? uint64
| +--ro one-way-utilized-bandwidth? uint64
| +--ro one-way-min-delay? uint32
| +--ro one-way-max-delay? uint32
| +--ro one-way-delay-variation? uint32
| +--ro one-way-packet-loss? decimal64
+--:(multi-nrp)
+--ro link-nrp-list
+--ro nrp* [nrp-id]
+--ro nrp-id uint32
+--ro nrp-link-attributes
+--ro link-partition-type? identityref
+--ro bandwidth-value? uint64
+--ro selector-id
| +--ro srv6? srv6-types:srv6-sid
| +--ro mpls-nrp-selector? uint32
+--ro statistics
+--ro status
| +--ro admin-status
| | +--ro status? identityref
| | +--ro last-change? yang:date-and-time
| +--ro oper-status
| +--ro status? identityref
| +--ro last-change? yang:date-and-time
+--ro one-way-available-bandwidth? uint64
+--ro one-way-utilized-bandwidth? uint64
+--ro one-way-min-delay? uint32
+--ro one-way-max-delay? uint32
+--ro one-way-delay-variation? uint32
+--ro one-way-packet-loss? decimal64
]]> shows the full tree diagram of the
NRPs device YANG model defined in module "ietf-nrp-device.yang".