Cisco Validated Blueprint

Architecture
For Building 5G Network Across DC and Transport

Sonu Kumar Khandelwal, Principal Technical Marketing Engineer

Shahid Ajmeri, Principal Product Manager
Session Objective

This session will cover the requirements and solutions of DC and

transport integrations. The goal would be to present and
demonstrate the design and products Cisco has validated in the
internal labs.

This session would focus on the integration of DC and Transport.

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 3
 • Importance of Converged DC and Transport
Architecture for 5G Advanced and 6G
• Requirements: DC and Transport
• Solution: Distributed DC and Transport
Integration
• Cisco Validated Design for DC to transport
Handoff with VXLAN EVPN and ACI

Agenda • Control plane, and data plane

• Configuration
• Demo

• Network slice automation and assurance

• Summary

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 4
Importance of
Converged DC and
Transport Architecture
for 5G Advanced and
6G
 5G | Buying Experience

4G | Buying Connectivity
© 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public
5G Business Enablers

Quality of Mobile video will

Experience FWA has emerged
continue to drive
Is more important as the first big 5G
the bandwidth
for consumers use-case
growth

5G
Enablers & Monetize 5G
Trends Edge-computing investments for New set of
is essential for enterprise consumer end-
latency-sensitive services – devices will drive
applications Private 5G, 5G connections
Network slicing scale

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 7
 Network Slice Expectations
Slice Monitoring and Service Assurance
Application endpoints are deployed in DC, while traffic from user to application is via transport. This creates a need for
end-to-end network slicing, assurance, and orchestration across DC and transport.

DC Switching DC Switching DC Switching

B C

A Fronthaul Midhaul Backhaul To Peering / Internet

1 End to End Service Definition Slice definition requires service guarantee between
Services are defined between service end points two service end points.
A <->B or B<—>C
Service Assurance is to ensure infra meets
2 Service Level Assurance application performance SLAs.

End-to-end slicing and assurance cannot be

3 Orchestration across domains delivered without orchestration across domains

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 8
Network Slice
End to End Network Slice is a construct of multiple slice instances, across various
service demarcation points, with common service level expectation.

RAN Slice Network Slice Core Slice

RAN Slice Network Slice Controller 5GC Slice

Controller Controller

Slice O-DU Slice O-CU Slice Slice

Instance 1 Instance 2 Instance 3 Instance 4

xHaul xHaul DC Handoff To App / Internet

DC Fabric 5GC NFs

End to End Network Slice

“Network Slice” includes Infrastructure (Transport

and DC) between Radio and 5GC functions

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 9
 • Ensure end to end service
quality
Importance of
Converged DC and • Automate, Orchestrate and
Transport Monitor Service performance

Architecture
• Enable growth opportunities
with Network slice as a
Service model

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 10
Requirements:
DC and Transport
Integration
DC requirements

Automation of
E<->W ECMP scale since
distributed DC
NF movement and communication single service is being
configuration across
flexible placement between application handled by multiple
100s of fabric and
endpoints smaller K8 Pods
1000s of switches

Integration with VMM

Horizontal scale for
domains (K8, End point visibility into
Service chaining endpoints, bandwidth,
OpenStack, vCenter fabrics
routing etc.
etc.)

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 12
Transport requirements

Traffic engineering Quality of service Vertical routing scale

Multi-vendor
automation and
Fast reroute (50 Multi-vendor SP core
operation tool to
msec convergence) network
enable end-to-end
network slicing

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 13
Handoff for best of both worlds
Datacenter Transport
VXLAN-EVPN/ACI SR-MPLS/SRv6

✓ Default standard for DC deployments

✓ Mature automation and operations tools for
DC use-cases Handoff ✓ Traffic engineering
✓ Distributed DC management ✓ Quality of service
✓ Mature and proven service chaining use- ✓ Vertical routing scale
cases ✓ Fast reroute (50 msec
✓ Distributed routing, forwarding and horizontal convergence)
scale for CNFs & VNFs ✓ Multi-vendor SP core network
✓ Flexible placement and Mobility of Workloads

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 14
Solution:
Distributed DC and
Transport Integration
Telco DC fabric architecture
Automation and operations across distributed fabrics

Orchestrator Cisco Nexus Dashboard Insights

APIC NDFC

ACI Remote Leafs NX-OS vPC

NX-OS vPC
ACI Fabric NX-OS Fabric Cloud network
controller

APIC

ACI Remote Leafs

1000+ Locations ACI Fabric NX-OS Fabric

Micro-second 100+ locations ACI Multi-Pod Fabric
latency sensitive Latency sensitive apps: UPF, Ent. Apps, vBNG, 2-10+ locations
apps: O/vDU Gi-LAN, IOT, Caching Non-Latency sensitive apps: control plane, IMS
Public Cloud
Far Edge DC Edge DC Central DC

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 16
 5G Converged SDN Transport Architecture
Cisco Automation & Service Assurance
CSR
FH
End-to-End IP with Segment Routing
4G, 5G

CSR Public Cloud

WiFi 6/7 MH
DC Fabric DC Fabric DC Fabric

DU
FWA / Premium
Services xHaul
vCSR MH

Enterprise Pre-Agg Aggregation Core

DU
VPNs vRouter
CSR
Cloud Services
PON BH
Cable
Legacy TDM
Cloud Ready Converged Architecture
DU
CU

BENEFITS

Optimized for Multi-Service Flexible and Precise network-based Seamless integration Consistent Open, automated,
Environments programmable timing and across Transport, DC QoS and Service centralized
(4G, 5G, FWA, Enterprise, architecture synchronization and Cloud Performance management at scale
PON and Legacy networks)

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 17
Benefits of SR handoff between DC and Transport

Network slicing across DC and

Unified SR based transport
transport

Scalable connectivity Cross-domain visibility

Scalable handoff for multiple VRFs for IOT,
Inter-DC user traffic visibility for transport team
enterprise 5G, private cloud use-cases

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 18
End-to-End
Network Slicing
with SR handoff
 How do I create Network slice?

NDFC

DC to transport handoff
APIC L3 VPN
VPNv4/v6

VRF-1 Border PE
VRF-2 DC - PE Metro Transport
DC
5GC
VRF-n
Transport
For creating fully automated and
DC end-to-end network slicing Transport network requires
across DC and transport, the network slicing due to
• DC fabric have large bandwidth, following parameters can be heterogeneous architecture
same/low latency, and high ECMP due used: and long-distance network.
to Clos architecture, hence network • BGP color community
slicing within DC fabric is typically not • VRF
required. • QOS - DSCP value
• EXP values
• However, DC applications (for • IPv4/IPv6 prefix
example 5GC functions) should signal
slicing in the transport network.

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 20
Slice configuration across DC and transport

DC specific config DC to transport handoff Transport specific config

• Tenant / VRF • SR handoff (Day-0) • ODN templates for SR policies

• Server Port Connectivity • Mark BGP Color Community, EXP, DSCP • L2VPN, L3VPN Services
• Network / VLAN / SVI for creating SR Policies in transport for • SR-MPLS, SRv6 config
• Service Chaining network slicing (Day 1)

NDFC

DC to transport handoff
APIC L3 VPN
VPNv4/v6

VRF-1 Border PE
DC - PE Metro Transport
DC
5GC
VRF-2 Transport
VRF-n

DC
BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 21
 Network slice using BGP color community
2

Mark DC app subnets with

BGP color community on
DC-Fabric “A” Border Leaf (BL) / Border
DC-Fabric “B”
PE (BPE)

BL/BPE BL/BPE
10.1.1.0/24 BL/BPE BL/BPE 1 10.1.2.0/24
App-1 subnet App-1 subnet
Define on-demand next-
hop (ODN) template based
3 on BGP color community
3
on DC-PE
SR tunnels are SR tunnels are
automatically created automatically created
based on incoming BGP DC PE DC PE DC PE
based on incoming BGP
color community and SR DC PE
color community and SR
ODN policy SP Transport ODN policy

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 22
Cisco Validated Design
for DC to transport
Handoff with :
VXLAN EVPN
VXLAN EVPN to transport handoff with SR-MPLS
• Single control plane and data plane session instead of per VRF control plane and sub-interface
• Unified SR based transport network

Single BGP session for all

VRFs
VXLAN Fabric
SP Transport
BGP AS-100
SR-MPLS SR-MPLS
NDFC

VXLAN + BGP EVPN MP-BGP L3 VPN

VPNv4/v6

VRF-1 Border PE PE
DC-PE
VRF-2
VRF-n BGP-AS 200
Seamless Integration of EVPN
with L3VPN (VPNv4/v6)

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 24
 VXLAN EVPN to SR-MPLS handoff control plane
VXLAN Fabric
BGP AS-100
• Border PE (BPE) is configured as a L3 VPN NDFC

(BGP VPNv4/v6 peer) with DC-PE to advertise VXLAN + BGP EVPN

overlay (VPN) SR label VRF-1 VRF-1

VRF-2 VRF-2
Server Leaf
VRF-n VRF-n
Border PE
• OSPF/ISIS/BGP-LU (BGP IPv4 labeled unicast) BGP AS-100
Border PE
VRF-1 BGP AS-100
can used between BPE and its neighboring VRF-2
NF
device to advertise underlay SR label VRF-1 VRF-n
(192.168.1.0/24)

• BPE receives the subnet (192.168.1.0/24) in

the BGP EVPN address-family, and imports OSPF/ISIS/BGP-LU
this prefix into VPNv4 address-family and L3 VPN (BGP VPNv4/VPNv6)
advertises to DC-PE SR-MPLS dataplane

• BPE receives the subnet (192.168.2.0/24) in

the BGP VPNv4 address-family, and imports DC-PE DC-PE
this prefix into EVPN address-family and BGP AS-200 BGP AS-200

advertises to server leaf SR-MPLS transport

External subnet
VRF-1 VRF-1 (192.168.2.0/24)
VRF-2 PE
VRF-n BGP AS-200

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 25
VXLAN EVPN to SR-MPLS handoff data plane (1)
Traffic from DC to transport
2
• BPE de-encapsulate VXLAN packet, identifies the VRF (VRF-1) based
1 on incoming VXLAN header

Server leaf encapsulates packets to the • BPE looks up the destination prefix into the VRF and encapsulate
external destination (192.168.2.0/24) into packet with SR-MPLS header and SR-MPLS VPN label.
VXLAN header and forwards to BPE

VXLAN Fabric
BGP AS-100
VXLAN + BGP EVPN
NDFC

VRF-1 VRF-1
VRF-2 VRF-2
Server Leaf VRF-n VRF-n
Border PE Border PE
BGP AS-100 BGP AS-100
VRF-1
NF
VRF-2
VRF-1 VRF-n SR-MPLS dataplane
(192.168.1.0/24)

DC-PE DC-PE
BGP AS-200 BGP AS-200

SR-MPLS transport External subnet

VRF-1 (192.168.2.0/24)

VRF-1
VRF-2
PE
VRF-n
BGP AS-200

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 26
VXLAN EVPN to SR-MPLS handoff data plane (2)
Traffic from DC to transport

VXLAN Fabric
BGP AS-100
NDFC
VXLAN + BGP EVPN

VRF-1 VRF-1
VRF-2 VRF-2
Server Leaf
VRF-n VRF-n
Border PE
BGP AS-100 Border PE
VRF-1 BGP AS-100
NF
VRF-2
VRF-1 VRF-n
(192.168.1.0/24) SR-MPLS dataplane

DC-PE DC-PE 3
BGP AS-200 BGP AS-200
• DC-PE de-encapsulate SR-MPLS packet. It
checks VPN label and identifies the VRF based on
SR-MPLS transport VPN label

VRF-1 • DC-PE does a look up for prefix in the VRF and

VRF-2 PE External subnet forwards packets to external destinations. It can
VRF-n BGP AS-200 VRF-1 (192.168.2.0/24)
apply SR policy in the transport domain as per
application requirement

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 27
Demo time – (VXLAN
EVPN to SR-MPLS
handoff with NDFC)
© 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public
SR-MPLS handoff with NDFC
NDFC manages both handoff and fabric configuration of both fabrics (DC1-VXLAN and DC2-VXLAN)
DC1-VXLAN
DC2-VXLAN

Site1-BPE-1
Site1-Leaf-1

Lo: 10.101.0.1
NDFC

Site2-BPE-1
Lo: 10.102.0.1
SID: 101

SID: 106
Lo: 201.1.1.1 Lo: 201.1.1.3
SID: 103 SID: 105

DC PE-1 DC PE-3
ISIS Process ISIS Process ISIS Process
(DC-Handoff) (CORE) (DC-Handoff)

DC PE-2
Lo: 201.1.1.2
VRF: sr_mpls_vrf_site2
VRF: DC-VRF1 Network: App2_site2:10.1.2.0/24
Site1-Leaf-2

SID: 104
Site1-BPE-2
Lo: 10.101.0.2

SP Transport
SID: 102

Topology diagram with ISIS and SR information

VRF: sr_mpls_vrf_site1
Network: App1_site1:10.1.1.0/24

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 32
Cisco Validated Design
for DC to transport
Handoff with :
ACI
 ACI to transport handoff using SR/MPLS handoff
• Single control plane and data plane session instead of per VRF control plane and data plane session
• Unified SR based transport network

Single BGP-EVPN session SR-MPLS

APIC for all VRFs

MP-BGP EVPN 192.168.1.0/24

Server Leaf
Prefix+Color

VRF-1 Border Leaf DC-PE

SP Transport
VRF-2
BGP-LU PE
NF
VRF-1 VRF-n
(192.168.1.0/24)
SR-MPLS Handoff

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 45
 ACI to SR-MPLS handoff control plane
• Border Leaf (BL) is configured as BGP
EVPN peering with DC-PE

• BGP-3107 (BGP IPv4 labeled unicast) is

used between BL and its neighboring APIC

device to advertise underlay SR label

Server Leaf
Border Leaf
• ACI fabric uses BGP VPNv4/v6 address BGP AS-100
Border Leaf
VRF-1 BGP AS-100
family within fabric VRF-2
NF
VRF-1 VRF-n
(192.168.1.0/24)
• ACI BL receives the subnet
(192.168.1.0/24) in the BGP VPNv4
address-family, and imports this prefix
eBGP IPv4 labeled unicast
into EVPN address-family and advertises BGP EVPN
to DC-PE SR-MPLS dataplane

• ACI BL receives the subnet

VRF-1 VRF-1
(192.168.2.0/24) in the BGP EVPN VRF-2 DC-PE DC-PE
BGP AS-200 BGP AS-200 VRF-2
address-family, and imports this prefix VRF-n VRF-n
into VPNv4 address-family and advertises SR-MPLS transport

to server leaf External prefix

PE VRF-1 (192.168.2.0/24)
BGP AS-200

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 46
 ACI to SR-MPLS handoff data plane (1)
Traffic from DC to transport 2
• BL de-encapsulate VXLAN packet, identifies the VRF (VRF-1)

• BL looks up the destination prefix into the VRF and encapsulate packet with SR-MPLS
1 header and SR-MPLS VPN label
Server leaf encapsulates packets to the
external destination (192.168.2.0/24) into
VXLAN header and forwards to BL

APIC

Server Leaf
Border Leaf Border Leaf
VRF-1 BGP AS-100 BGP AS-100
NF
VRF-2
VRF-1 VRF-n
(192.168.1.0/24)

VRF-1 VRF-1
DC-PE DC-PE
VRF-2
BGP AS-200 BGP AS-200 VRF-2
VRF-n VRF-n
SR-MPLS transport

DC-PE External prefix

192.168.1.0/24
VRF-1 (192.168.2.0/24)

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 47
 ACI to SR-MPLS handoff data plane (2)
Traffic from DC to transport

APIC

Server Leaf
Border Leaf Border Leaf
VRF-1 BGP AS-100 BGP AS-100
NF
VRF-2
VRF-1 VRF-n
(192.168.1.0/24)

3
VRF-1
DC-PE DC-PE • DC-PE de-encapsulate SR-MPLS packet. It
VRF-2
BGP AS-200 BGP AS-200
VRF-n checks VPN label and identifies the VRF based on
SR-MPLS transport VPN label

DC-PE • DC-PE does a look up for prefix in the VRF and

VRF-1 forwards packets to external destinations. It can
VRF-2 External prefix apply SR policy in the transport domain as per
VRF-n VRF-1 (192.168.2.0/24)
application requirement

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 48
ACI to SR-MPLS
handoff demo
© 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public
 Topology
APIC manages both handoff and fabric configuration

ACI Fabric
SP Transport
ACI Fabric
SP Transport

Lo: 14.1.1.1
Leaf104
BGP AS-65000
BGP AS-100

SID: 104
APIC

Lo: 5.5.5.5 Lo: 201.1.1.3

SID: 105 SID: 106

DC-PE-1 PE-2
Infra Tenant ISIS Process
SR-MPLS-Infra-L3out (CORE)
BGP-LU + BGP-EVPN
VRF: vrf1
Network: 26.1.1.0/24
Network: 8.8.8.8

Tenant: CiscoLiveAMS
Leaf 103

VRF: vrf1
Lo: 13.1.1.1

EPG/BD/Subnet: epg1/bd1/25.1.1.0/24
SID: 103

EPG/BD/Subnet: epg1/bd1/25.2.1.0/24
SR-MPLS-VRF-L3out: SR-MPLS-VRF-L3out
Topology diagram ACI and SR-MPLS handoff

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 53
Automating
End-to-End
Network Slicing
 Requirements to Automate Slice Across Domains
CNC Slice Controller
OR

?
NDFC APIC NSO

• Tenant / VRF • SR Handoff (Day-0) • ODN templates for SR policies

• Server Port Connectivity • Mark BGP Color Community, EXP, • L2VPN, L3VPN Services
• Network / VLAN / SVI DSCP for creating SR Policies in • SR Configurations
• Service Chaining transport for network slicing (Day 1)

NDFC

DC to Transport Handoff
APIC L3 VPN
VPNv4/v6

VRF-1 Border PE
DC - PE Metro Transport
VRF-2 Transport
DC VRF-n

DC
BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 74
 Network Slice Automation: DC and Transport

• DC Specific configuration is
CNC Slice Controller
done by APIC/NDFC OR
APIC NDFC NSO
• Transport configuration is
done by CNC through NSO Instruct DC controller
to provision BL/BPE
for slicing [*Roadmap]
• Network Slicing across
domains are provisioned by
CNC for Day 1
• Transport is done by NDFC

NSO
• DC is instructed to APIC

L3 VPN
provision BL/BPE via VPNv4/v6
Border PE
APIC/NDFC VRF-1
VRF-2 Metro Transport
DC VRF-n
DC - PE
Transport
DC
BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 75
 Available in CNC 6.0

Transport Slicing Automation with CNC

Slice Designer Persona Slice Requestor Persona
(workflow done once) (workflow done 1000s of times)

Deploy the Pre-Requisites: Request a new Slice Instance :

SR-MPLS or SRv6, QoS, ODN, etc. • Pick Intent from catalog, select endpoints, Submit!
Build the Slice Catalog of Intents

Input

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 76
 Available in CNC 6.0

Simplified Transport Slice Creation- 4 Steps!

Specify Slice ID, Service Specify Slice Intent from Specify Connectivity Define Service
Type, Customer, Template Catalog Type, Isolation, Demarcation Points
Description,
Bandwidth… (endpoints)
NSSAI (optional)…

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 77
 Available in CNC 6.0

Navigate the Slice components: VPN, Transport

Display a slice on the map

View Slice and VPN view From the VPN list, display From the Transport list,
Visualize Shared
along with Shared Slices VPN details including display SR TE details
Slices associated to
and CE ( Neighbor) Assurance data if including SR-PM data if
dedicated slice
connected in Logical View monitoring is enabled SR-PM is enabled

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 78
Transport slice Demo
© 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public
Extending Transport Slice of CNC into DC
Create Network Slice across DC and Transport
Configure Hand-Off
A) Configure handoff parameters for prefixes
L3 VPN
advertised from DC fabric
VPNv4/v6 o For example: BGP color community
DC Transport

Auto provision Transport Slice B) Use existing forwarding plane policy (ODN
policy) to create slice in transport based on
L3 VPN BGP color community
VPNv4/v6

DC Transport

Monitor DC-BL as Managed CE of Transport

C) Create visualization for DC-BL as a CE into
CNC
o Extend Health assurance of slide across DC
L3 VPN
VPNv4/v6
and Transport
Managed CE o Monitor control plane health check – BGP
DC Transport Peering, BFD neighbors, interface status, IGP
protocol health (BGP-LU, ISIS, OSPF etc)

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 81
Evolution of network slicing model of CNC
1 2 3

Slicing service Slicing topology

Slicing catalog (modifications in existing (new model to support DC to
(no change) slice service model to transport handoff to include DC-
include DC topology) BL and DC-PE topology)

Topology options for slicing model

Topology Model

Inter-as option A (VRF-lite) Available (CNC 6.0)

ACI to SR-MPLS handoff Initial definition complete

VXLAN to SR-MPLS handoff Initial definition complete

ACI/VXLAN to SRv6 handoff Planned

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 82
 FYI

Use Case 1: VXLAN to SR-MPLS handoff

Yang model to define DC and transport Topology details

module: cisco-network-slice-service
+--rw slice-topology*
| +--rw sap-id* [name] e.g. Interconnect-1
| | +--rw connection-type? enumeration { VXLAN-SR-MPLS-L3VPN }
| | +--rw interconnect* [name]
| | +--rw dc-end ## List of VXLAN BL ##
| | | +--rw dc-controller [id] --> /slice-topology/node-settings/dc-controller[id]
| | | +--rw border-leaf [id] -> /devices/device:name=current()/../../..//leaf-switch/name
| | +--rw pe-end ## List of DC-PE ##
| | | +--rw pe-device --> /slice-topology/node-settings/pe-device[id]

SAP- Service access point (defined as per IEFT northbound interface yang for network slice)
https://datatracker.ietf.org/doc/draft-ietf-teas-ietf-network-slice-nbi-yang/
Disclaimer: Information in this example subject
to change with actual deployment on product

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 83
 FYI

Use Case 1: VXLAN to SR-MPLS handoff

Yang model to define slice service

module: cisco-network-slice-services
+--rw slice-service*
| +--rw sap-id [id]
| | +----sap-peering peer-sap-id [name] - interconnect-1 from slice-topology/sap-id
Flag: - Provision DCI - Yes/No [NO] —> if YES, get slice-topology/sap-id- interconnect/connection-type
| | | | +--rw dc-Tenant ( optional ) ---> /devices/device:name=current()/../../..//Tenant/vrf ## User to pick vrf that exists in leaf
switch , if not picked then slice service creates vrf ###
| | | | +--rw dc-prefixes [id] (optional) - one or more comma separated prefix address . Auto create Prefix list based on these
prefixes.
| | | | +--rw dc-outbound-route-map ( optional ) ---> /devices/device:name=current()/../../..//leaf-switch/name/route-map ##
User to pick route-map that exists in leaf switch , if not picked then slice service creates route-map ###

SAP- Service access point (defined as per IEFT northbound interface yang for network slice)
https://datatracker.ietf.org/doc/draft-ietf-teas-ietf-network-slice-nbi-yang/ Disclaimer: Information in this example subject
to change with actual deployment on product

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 84
 FYI

Use Case 2: ACI to SR-MPLS handoff

Yang model to define DC and transport Topology details

module: cisco-network-slice-service
+--rw slice-topology*
| +--rw sap-id* [name] e.g. Interconnect-1
| | +--rw connection-type? enumeration { ACI-SR-MPLS}
| | +--rw interconnect* [name] | | +--rw dc-end
| | | +--rw dc-controller [id] --> /slice-topology/node-settings/dc-controller[id]
| | | +--rw SR-MPLS-Infra-L3out [string] ---> /devices/device:name=current()/../../../devices/networking/sr-mpls-infra-l3out
### User to pick sr-mpls-l3out that exists in APIC controller
| | +--rw pe-end
| | | +--rw pe-device [id] --> /slice-topology/node-settings/pe-device[id]

SAP- Service access point (defined as per IEFT northbound interface yang for network slice)
https://datatracker.ietf.org/doc/draft-ietf-teas-ietf-network-slice-nbi-yang/ Disclaimer: Information in this example subject
to change with actual deployment on product

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 85
 FYI

Use Case 2: ACI to SR-MPLS handoff

Yang model to define slice service

module: cisco-network-slice-services
+--rw slice-service*
| +--rw sap-id [id]
| | +----sap-peering peer-sap-id [name] - interconnect-1 from slice-topology/sap-id
Flag: - Provision DCI - Yes/No [NO] —> if YES, get slice-topology/sap-id- interconnect/connection-type
| | | | +--rw dc-Tenant ( optional ) ---> /devices/device:name=current()/../../..//Tenant/vrf ## User to pick vrf that exists in leaf
switch , if not picked then slice service createsTenant and vrf ###
| | | | +--rw dc-prefixes [id] (optional) - one or more comma separated prefix address . Auto create Prefix list based on these
prefixes.
| | | | +--rw dc-outbound-route-map ( optional ) ---> /devices/device:name=current()/../../..//leaf-switch/name/route-map ##
User to pick route-map that exists in leaf switch , if not picked then slice service creates route-map ###

SAP- Service access point (defined as per IEFT northbound interface yang for network slice)
https://datatracker.ietf.org/doc/draft-ietf-teas-ietf-network-slice-nbi-yang/ Disclaimer: Information in this example subject
to change with actual deployment on product

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 86
Summary: End to End Slicing
Validated scalable handoff with automation and assurance

1. Validated Design options for seamless and scalable protocol

level hand-off between “Data Center” and “Transport”
infrastructure
DC Transport

2. Enabling capabilities to support End to End Slicing

provisioning up to service end-points
o O-RU to O-DU
o O-DU to O-CU to UPF etc.
DC Transport

3. Monitors and assures handoff connectivity between DC

and transport
B
A

DC Transport

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 87
Learning Map
Feb 5 | 2:10 pm Feb 8 | 8:45 am
PSOSP-1725 BRKSP-2189
How Automated Assurance Improves Cisco Validated Blueprint Architecture for
Service Quality of Experiences Building 5G Networks across DC and
Transport
Feb 6 | 9:30 am
Feb 8 | 3:00 pm
BRKSPG-2133
Evolution of the Transport Network BRKSPG-2263
Architecture for 5G and Beyond Design, Deploy and Manage Transport
Slices using SDN Controller and Assurance
Feb 7, | 10:45 am
BRKSPG-2315
Cloud-Ready Converged SDN Transport

Feb 8 | 8:30 am
LTRSP-2119
Hybrid Cloud Networking using Cisco XRd

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 88
Reference Links and Materials
Architecting the Telco Data Center with Cisco NX-OS and NDFC white paper

Cisco ACI Design Guide for Telco Data Center Deployments

Validated Design for Cisco ACI to SR-MPLS Handoff

ACI to SR-MPLS whitepaper

Converged SDN transport

BRKSP-2189 © 2024 Cisco and/or its affiliates. All rights reserved. Cisco Public 89
Thank you