Juniper Validated Design (JVD)

5G Fronthaul Network Using Seamless MPLS

Segment Routing

Juniper Networks Validated Designs provide customers with a comprehensive, end-to-end blueprint
for deploying Juniper solutions in their network.

These designs are created by Juniper's expert engineers and tested to ensure they meet the
customer’s requirements.

Using a validated design, customers can reduce the risk of costly mistakes, save time and money, and
ensure that their network is optimized for maximum performance.
About this Document
This document provides a Juniper Validated Design (JVD) for a 5G Fronthaul network using Seamless Segment Routing
Multiprotocol Label Switching (SR-MPLS) with Juniper Network’s next-generation ACX7000 platform. Through careful
testing, we found that the ACX7100-32C/48L and ACX7509 are excellent choices for access and aggregation purposes.
They offer enhanced performance and a wide range of advanced features, outperforming previous ACX platforms in most
scenarios. Both the ACX7100 and ACX7509 are particularly well-suited for the Hub Site Router (HSR) or Lean Edge
segments, as they provide the necessary scale, bandwidth, feature velocity, and performance capabilities. While the
ACX7100 surpasses the requirements for access nodes, it’s also an ideal option for 400G Fronthaul or Metro Access
deployments.

Documentation Feedback
We encourage you to provide feedback so that we can improve our documentation. Send your comments to design-
center-comments@juniper.net. Include the document or topic name, URL or page number, and software version (if
applicable).

Table of Contents
Solution Benefits 3
Use Case and Reference Architecture 4
Solution Architecture and Design 6
Overlay Services 7
Fronthaul Network Design and Topology 8
Fronthaul Layer 2 Connectivity Models 8
Layer 3 Connectivity Models 10
VLAN Operations 11
Solution and Validation Key Parameters 13
Supported Platforms and Positioning 13
Solution Design and Infrastructure 13
Scale and Performance 13
Anticipated Fronthaul Network Resiliency and Latency 14
Key Feature List 14
Test Bed 15
Solution Validation Goals 16
Solution Validation Non-Goals 17
Results Summary and Analysis 17
ECMP Load Balancing 17
Convergence Briefing 18
Recommendations 20
Solution Benefits
A 5G Fronthaul network using seamless MPLS segment routing offers several advantages:

• Low Latency: 5G networks require ultra-low latency to support real-time applications such as autonomous
vehicles, remote surgery, and virtual reality. ACX7000 Cloud Metro routers assure an ultra-low latency
forwarding while SR-MPLS provides an optimized latency based path forwarding in the network eliminating
the need for complex protocol processing.
• Scalability: The 5G network infrastructure needs to support a massive number of devices and provide
seamless connectivity.SR-MPLS enables network scalability by simplifying the forwarding plane and reducing
the control plane complexity. This allows for efficient resource utilization and optimized network performance.
• Traffic Engineering: SR-MPLS provides advanced traffic engineering capabilities, enabling operators to
dynamically control and optimize the flow of traffic. This allows for efficient load balancing, congestion
avoidance, and quality of service (QoS) management, which are essential for delivering high-performance 5G
services.
• Network Resiliency: 5G networks require high availability and resiliency to ensure uninterrupted service. SR-
MPLS offers fast rerouting mechanisms and supports protection and restoration schemes, allowing the
network to quickly recover from failures and maintain service continuity.
• Simplified Operations: SR-MPLS simplifies network operations by leveraging source routing. Instead of
maintaining complex routing tables at every network node, the forwarding path is explicitly encoded in the
 packet header. This simplifies network configuration, reduces control plane overhead, and improves overall
operational efficiency.

Use Case and Reference Architecture

The 5G xHaul architecture encompasses three physical segments: Fronthaul, Midhaul, and Backhaul as shown in Figure 1.

The primary focus of the JVD is on providing design options and performance validation of the Fronthaul network
considered in the context of the end-to-end converged 5G xHaul network.

Figure 1: 5G xHaul Reference Network

Fronthaul segment enables Layer 2 connectivity between Open Radio Unit (O-RU) (cell site) and Open Distributed Unit (O-
DU) (shown as RU and DU in Figure 1) in the radio access network (RAN), allowing them to communicate for control, data,
and management traffic. It also ensures time and frequency synchronization between RAN elements. Because low latency
is crucial (must be below 150µs from RU to DU), the Fronthaul segment has very few network elements, typically limited
to one or two hops. The current solution for 5G Fronthaul transport is based on O-RAN Alliance architecture [ORAN-
WG9.XPSAAS.0-v00.01].

The advancement of the RAN involves different architectures for 4G, including distributed, centralized, and virtual setups,
which need to coexist with the 5G disaggregated O-RAN. These diverse ecosystems provide flexibility for the placement of
components like O-DU and O-CU. This JVD does not cover all possible scenarios but closely aligns with O-RAN split 7.2x,
where the O-RU connects to the CSR and the O-DU is located within the HSR infrastructure. If needed, additional
insertion points can be implemented to support disaggregation between the Midhaul and Backhaul segments by extending
appropriate services.

Figure 2 (as proposed by the O-RAN Alliance), summarizes the deployment scenarios for the RAN according to the ITU-T
for simultaneous support of 4G and 5G.
 Figure 2: RAN Deployment Scenarios for Simultaneous Support of 4G and 5G

Figure 3 shows an end-to-end 5G xHaul network, modeled after common topology [O-RAN.WG9.XPSAAS-v02.00], which
defines four segments of transport infrastructure: access, pre-aggregation, aggregation, and transport core.

Figure 3: 5G xHaul Network Topology

Topology Definitions:

1. Access: ACX7100-48L as DUT (AN3), ACX7100-48L (AN1), ACX710 (AN2) cell site routers.
2. Pre-Aggregation: ACX7509 (AG1.1) and ACX7100-32C (AG1.2) DUT hub site routers.
3. Aggregation: MX204s (AG2.1/AG2/2), MX10003 (AG3.1), MX480 (AG3.2) aggregation routers.
4. Core Network: PTX1000 (CR1) and MX10003 (CR2) core routers. MX10003 (SAG) services router.

The network topology in this use case combines Seamless MPLS and ISIS Segment Routing technologies. These
technologies enable the support of multiple 5G and 4G services, such as:

• Layer 3 VPN (IPv4 and IPv6)

• Ethernet VPN–virtual private wireless service (EVPN-VPWS)
• EVPN Flexible Cross Connect (EVPN-FXC)
• L2VPN
• L2Circuit
• Border Gateway Protocol-Virtual Private LAN Service (BGP-VPLS)

Solution Architecture and Design

The reference solution architecture for end-to-end xHaul, shown in Figure 4, deploys a spine-leaf access topology
(Fronthaul) and core, peering and services gateway roles comprising the xHaul infrastructure.

The network setup includes Segment Routing MPLS as the underlying technology, which spans across multiple ISIS
domains and inter-AS connections. The access nodes are placed within an ISIS L1 domain, establishing adjacencies with
L1/L2 HSR nodes. The L2 domain extends from the aggregation to the core segments of the network. To achieve seamless
MPLS connectivity, BGP-Labeled Unicast (BGP-LU) is enabled at the border nodes.
 Figure 4: JVD 5G Fronthaul Services Topology

To manage current network scale and allow future growth, two sets of route reflectors are used at CR1 and CR2, primarily
serving the westward HSR (AG1) clients. AG1.1/AG1.2 act as redundant route reflectors specifically for the access
Fronthaul segment. Inter-AS Option-B solutions are supported through Multi-Protocol BGP peering between the Services
Aggregation Gateway router (SAG) and the HSR (AG1).

To ensure the performance and reliability of the network, the design proposes additional enhancements, including E-OAM
performance monitoring and Flow-Aware Transport Pseudowire (FAT-PW) load-balancing. These enhancements help
monitor the network performance and balance the traffic load efficiently.

Overlay Services
The overlay services in the network use different combinations of VLAN operations. These operations are applied to each
Layer 2 service types, including EVPN, L2Circuit, VPLS, and L2VPN. Additionally, L3VPN services incorporate IPv6
tunnelling to validate 6vPE functionality.
As shown in Figure 4, inter-domain VPN Option-B is used at the HSR/AG1 region, and Inter-AS Option-C is used between
the SAG and CR border nodes. These nodes are enabled by BGP-LU.

This combination of VPN’s is designed in a way to allow following traffic flows in the 5G xHaul network:

• Layer 2 eCPRI between O-RU to O-DU traffic flows – 5G Fronthaul

• Layer 3 IP packet flows between 5G O-DU and CU/EPC – 5G Midhaul and Backhaul
• Layer 3 IP packet flows between 4G CSR and EPC (SAG) – 4G L3-MBH
• Layer 2 flows between CSR (AN) to EPC (SAG) – 4G L2-MBH
• Layer 2 Midhaul flows emulating additional attachment segments – 4G Midhaul and Backhaul
For more detailed information about this architecture, contact your Juniper Networks representative.

Fronthaul Network Design and Topology

The Fronthaul network deployment scenarios were carefully designed to support both the traditional 4G mobile backhaul
and the evolution into the 5G network infrastructure over the same physical network. This approach allows MSOs to make
a smooth transition from 4G to 5G without disrupting their existing services. They can gradually introduce the necessary
changes and upgrades to accommodate the new requirements of 5G networks.

As shown in Figure 5, the Fronthaul network consists of ACX7000 series routers interconnected by high-capacity links.
The ACX7100-48L serves as the CSR, providing connectivity between O-RU and HSR aspects of the RAN. The ACX7100-
48L supports a range of port densities, including 47 ports of 10/25/50G + 1x10/25G and 6 x 400G (24x100G). The ability
to support 400G access topologies is a key building block to the solution.

Figure 5: ACX Platforms Positioning Within Fronthaul Network Topology

Fronthaul Layer 2 Connectivity Models

There are several possible connectivity models between O-RU and O-DU, but for the purposes of this JVD, we’ve
validated these two:

• EVPN-VPWS Single-Homed supporting dedicated MAC for eCPRI without redundancy

• EVPN-VPWS with A/A LAG DU attachment
Figure 6 illustrates the first connectivity model. In this scenario, the network utilizes EVPN-VPWS single-homing
connectivity. This setup supports dedicated MAC for eCPRI without redundancy. Additionally, it uses Ethernet OAM with
performance monitoring. Currently, Ethernet OAM with performance monitoring is supported for the single-homed
configuration in this model.

Figure 6: EVPN-VPWS Single-Homed Supporting Dedicated MAC for eCPRI Without Redundancy

Figure 7 illustrates the second connectivity model. In this scenario, the network uses two configurations between the CSR
(AN3) and HSR-1/HSR 2:

• EVPN-VPWS with active-active multihoming

• EVPN-VPWS with Flexible Cross Connect (FXC) active-active multihoming
To enable traffic load sharing, an Active/Active Ethernet Segment Identifier (ESI) Link Aggregation Group (LAG) is
established between the HSRs and the O-DU. This allows for balanced distribution of traffic. The links are bundled into an
Active/Active EVPN ESI 10G Ethernet LAG between HSR-1 and HSR-2, as well as to the O-DU. The O-DU includes a two-
member Aggregate Ethernet (AE) with both links actively carrying traffic.

In this scenario, eCPRI packets might arrive on either of the O-DU links from the HSRs. Similarly, eCPRI packets can be
sent across either of the HSR-1 or HSR-2 uplinks to support active-active functionality. This setup ensures flexibility and
redundancy in the network for improved performance.

Figure 7: O-RAN Fronthaul Network A/A EVPN-VPWS

Layer 3 Connectivity Models
We chose L3VPN protocol to facilitate Layer 3 connectivity between O-DU and vCU/vEPC elements of the 5G xHaul. The
JVD proposes two different connectivity models between the Hub Site Router (HSR) and the O-DU, with both supporting
L3 multihoming between O-DU and pair of HSRs:

• EVPN IRB anycast gateway with L3VPN, refer to Figure 8

• BD with IRB and static MAC/ARP with L3VPN, refer to Figure 9
We tested both connectivity models independently and collected data on their coexistence and convergence. We
validated that both models are effective in facilitating the necessary connectivity between the HSR and O-DU.
For further details about the testing for these connectivity models, contact your Juniper Networks representative.

Figure 8: EVPN IRB Anycast Gateway with L3VPN

Figure 9: BD with IRB and Static MAC/ARP with L3VPN

VLAN Operations
The ACX7000 series offers a wide range of VLAN operations. The proposed VLAN operations specifically cater to EVPN
eCPRI-emulated traffic flows and support various normalization scenarios:

• Untagged frames
• Tagged (802.1q)
• Dual-tagged (802.1ad)
While the main emphasis of this JVD is on the Fronthaul segment, we also validated most of the VLAN operations for each
VPN service type. To see the full test report, contact your Juniper Networks representative.

Figure 10: Ethernet Encapsulation and VLAN Normalization of eCPRI Flows

Figure 11: Ethernet Encapsulation and VLAN Normalization of eCPRI Flows (cont’d)
Solution and Validation Key Parameters

Supported Platforms and Positioning

Table 1: Supported Platforms and Positioning
Platform Device Junos Release
DUT Platforms ACX7100-48L, ACX7100-32C, ACX7509 Junos OS Evolved 22.3R1-S1

Helper Platforms ACX710, MX480, PTX1000, MX10003 Junos OS 22.3R1

Solution Design and Infrastructure

• 5G xHaul MBH reference architecture
• Seamless MPLS across xHaul IGP domains (Inter-AS BGP-LU)
• Fast failover & detection TI-LFA, BFD, OAM, ECMP, etc.
• Segment Routing L-ISIS
• Redundant Route Reflectors
• EVPN-VPWS and FXC A/A Multihoming
• EVPN-ELAN with Multihoming and EVPN IRB
• BGP-VPLS, L2VPN, L2Circuit, L3VPN
• Inter-AS Option B/C
• O-RAN Alliance [ORAN.WG9.XPSAAS-v02.00]

Scale and Performance

This section contains key performance indexes (KPIs) used in solution validation targets. Validated KPIs are multi-
dimensional and reflect our observations in customer networks or reasonably represent solution capabilities. These
numbers do not indicate the maximum scale and performance of individual tested devices. For uni-dimensional
data on individual SKUs, kindly contact your Juniper Networks representatives.
The Juniper JVD team continuously strives to enhance solution capabilities. Consequently, solution KPIs may
change without prior notice. Always refer to the latest JVD test report for up-to-date solution KPIs. For the latest
comprehensive test report, please reach out to your Juniper Networks representative

Table 2: Scale Summary

AN3 (ACX7100-48L) AG1.1 (ACX7509) AG1.2 (ACX7100-32C) SAG (MX10003)
Feature
Access / CSR Pre-Agg / HSR Pre-Agg / HSR Services Agg
RIB/FIB 351k/123k 303k/121k 295k/121k 381k/116k

VLANS 2115 7300 7300 8414

EVPN-VPWS SH 1000 1000 1000 0

EVPN-VPWS MH 101 101 101 0

 AN3 (ACX7100-48L) AG1.1 (ACX7509) AG1.2 (ACX7100-32C) SAG (MX10003)
Feature
Access / CSR Pre-Agg / HSR Pre-Agg / HSR Services Agg
EVPN-ELAN (L3VPN) 0 500 500 0

Bridge Domain (L3VPN) 0 500 500 0

L2Circuit 200 1000 1000 2400

L2VPN 200 1000 1000 2410

L3VPN (OSPF) 100 100 100 1101

L3VPN (BGP) 100 100 100 100

VPLS 200 1000 1000 2400

MAC (VPLS) 10.6k 22k 102k 145k

CFM UP MEP 1200 1000 1000 1000

Anticipated Fronthaul Network Resiliency and Latency

The solution creates a resilient network system that safeguards against link or node problems in the Fronthaul segment. It
ensures that if a link fails between CSR and HSR nodes, the traffic flow will be restored within 50 milliseconds or less.
Additionally, the expected delay for packets between RU and DU should not exceed 120 microseconds.

Key Feature List

• EVPN-VPWS
• EVPN-ELAN
• EVPN-FXC
• L3VPN
• BGP-VPLS
• L2Circuit
• L2VPN
• Segment Routing ISIS
• TI-LFA (link/node)
• ISIS
• BGP
• BGP-LU
• BFD
• Community-based Routing Policy
• Route Reflection
• IPv4
• IPv6
• LACP
• AE
• CFM
• LFM
• VLAN (802.1q)
Contact your Juniper Networks representative for the full test report and feature list.

Test Bed
Figure 12 illustrates the test bed we used. The network consists of four layers: access, pre-aggregation, aggregation, and
transport core.

• Fronthaul segment: Uses a spine-leaf access topology, connecting to redundant HSR (AG1.1/1.2) nodes, which
also handle 4G pre-aggregation and 5G HSR functions. The pre-aggregation AG1 nodes provide connectivity
for O-DUs and include additional emulated access insertion points, specifically the Remote Tester (RT), for
scalability purposes.
• Midhaul and Backhaul segments: These are represented by ring topologies and serve aggregation and core
roles, but they are not the main focus of this discussion.
Figure 12: JVD 5G Fronthaul Lab Topology
Table 3: Topology Definitions
Layer Devices Under Test
Access ACX7100-48L as DUT (AN3), ACX7100-48L (AN1), ACX710 (AN2) cell site routers

Pre-Aggregation ACX7509 (AG1.1) and ACX7100-32C (AG1.2) DUT HSRs

Aggregation MX204s (AG2.1/AG2/2), MX10003 (AG3.1), MX480 (AG3.2) aggregation routers

Core Network PTX1000 (CR1) and MX10003 (CR2) core routers. MX10003 (SAG) services router

In the testing process, we used three main devices as Devices Under Test (DUT) to run all the test cases. The ACX7100-
48L router was used as the Access Node Cell Site Router (CSR) for both 5G and 4G. The ACX7100-32C and ACX7509
routers were used as the Pre-Aggregation Hub Site Routers (HSR) for the AG1 role.

To support and expand the testing environment, we introduced additional helper nodes. We added another ACX7100-48L
router (AN1) to scale the setup and emulate identical traffic flows between each HSR DUT, ensuring accurate and uniform
validation against both ACX7509 and ACX7100-32C HSRs. MX routers were used in different roles, such as Aggregation
(AG2/AG3), Core (CR2), and services router (SAG). The PTX platform was used for the Core Router (CR1).

For a complete test report with detailed information about the hardware and software used, contact your Juniper
Networks representative.

Solution Validation Goals

Here are the major test goals:

• Validate VPN services, including L3VPN, EVPN-VPWS, EVPN-FXC, EVPN-ELAN, BGP-VPLS, L2Circuit, and
L2VPN over SR-MPLS transport architecture.
• Validate TI-LFA redundancy mechanisms over Segment Routing with Seamless MPLS / BGP-LU
• Validate network resiliency, traffic restoration, and measured convergence time for ACX7100-48L (AN3),
ACX7100-32C (AG1.2) and ACX7509 (AG1.1), with adjacent link/node failures for all traffic types.
• Measure solution resilience of layer 2/3 flows from specified DUTs, including:
 Access Node (AN) to Pre-Aggregation AG1 (O-RU to O-DU)
 Pre-Aggregation AG1 (including O-DU) to Services Gateway (SAG)
 Access Node (AN) to SAG (EPC)
• Validate network stability for major 4G/5G traffic flows at scale with each VPN service type over SR-MPLS
during normal and stress conditions.
• Validate consistency and resiliency of the ACX7100-48L, ACX7100-32C and ACX7509 systems against
negative stress conditions (enabled/disable control and data plane daemons, add/delete configurations, etc.)
• Identify product limitations, anomalies, and open Problem Reports (PRs) exposed during validation stage.
• OAM Connectivity Fault Management (CFM), Y.1731over VPLS and EVPN-VPWS services.
• Flow-Aware Transport Pseudowire (FAT-PW) load-balancing.
Solution Validation Non-Goals
Non-goals represent protocols and technologies outside the scope of the current validation.

• Class of Service
• Underlay MPLS/SR transport other than specified in the Test Goals
• Network Slicing
• Flex Algo
• Classful Transport, BGP-CT, SSR
• BGP-PIC
• Management and Automation
• Chassis-based High Availability (as applicable)
• End-to-End Timing and Synchronization Distribution: Synchronous Ethernet, IEEE1588v2
• SLA Monitoring: RFC 2544, Y.1564, TWAMP, Active Assurance
• Telemetry

Results Summary and Analysis

Within the course of validation, we were able to demonstrate a robust solution for 5G Fronthaul and Midhaul transport
infrastructure utilizing Seamless MPLS. The JVD generates reasonable multi-vector scale of L2/L3 connectivity services as
compared with MNO and MAN operator expectations for real network deployments, while satisfying stringent SLA
requirements. The scale reference in Table 2 characterizes primary multidimensional KPI’s represented in the validated
profile.

ACX7100-48L, ACX7100-32C and ACX7509 as the DUT routers have each successfully passed 186 test cases curated to
support the given reference architecture.

With this network design, the architecture can deliver fast restoration within 50ms for most traffic flows transported over
ISIS-SR with TI-LFA. Load distribution and optimization knobs were shown to improve service restoration against link/node
failures. Link events consistently achieve <50ms convergence, see Table 5. Node failures were more disruptive as
expected, exacerbated by scale and in some rare cases, production limitations. Contact your Juniper Networks
representative for a complete results report and analysis.

ECMP Load Balancing

The design minimizes traffic issues when links or nodes fail by distributing the load across equal cost paths and making use
of TI-LFA protection mechanisms. Multiple load-sharing methodologies are incorporated, such as adjusting IGP metrics,
BGP multipath, ECMP fast-reroute, and VPN-unequal-cost for L3VPN services. We also activated FAT-PW label on
ACX7000 devices where supported, including Layer 2 VPN and Layer 2 Circuits, which enables additional granularity to the
hash computation. Hash-key configuration is mandatory on ACX7000 series to perform traffic load balancing. By default,
there are no hash-keys enabled. This validation enables all relevant hash-keys for extracting L2, L3 and MPLS fields.
The full test report includes details about the load sharing configurations enabled for this validation. Junos OS Evolved
22.3R1-S1/R2 and later are recommended versions for ACX7000 series and includes additional load sharing
enhancements. Contact your Juniper Networks representative if you’d like to receive the complete report.

Table 4: ACX7000 ECMP Summary Results

ACX7000 ECMP Load Balancing Performance [1]

Service: DUT (Traffic Path) ECMP Flow FAT-PW Link1 Link2 Link3 Link4
Links LAG (#) LAG (#) LAG (#) LAG (#)

L2VPN: ACX7100-48L (AN3 to SAG) 3 100kfps Y ae20(2) ae23(1) et-0/0/50 N/A

36.6kpps 33.1kpps 33.1kpps

L2CKT: ACX7100-48L (AN3 to SAG) 2 100kfps Y 49.2kpps et-0/0/50 N/A N/A

53.2kpps

VPLS: ACX7100-48L (AN3 to SAG) [2]

3 100kfps N ae20(2) ae23(1) et-0/0/50 N/A
36.0kpps 33.2kpps 33.7kpps

EVPN: ACX7100-48L (AN3 to DU) 3 100kfps N ae20(2) ae23(1) et-0/0/50 N/A

36.0kpps 33.4kpps 33.0kpps

L3VPN: ACX7100-48L (AN3 to AG1.2) 3 100kfps N ae20(2) ae23(1) et-0/0/50 N/A

36.9kpps 33.7kpps 33.4kpps

L2VPN: ACX7509 (AG1.1 to SAG) 2 100kfps Y ae4(2) ae5(1) N/A N/A

50kpps 49.6kpps

L2CKT: ACX7509 (AG1.1 to SAG) 2 100kfps Y ae4(2) ae5(2) N/A N/A

53.2kpps 50.2kpps

VPLS: ACX7509 (AG1.1 to SAG) [2] 2 100kfps N ae4(2) ae5(2) N/A N/A
53.3kpps 50.1kpps

EVPN: ACX7509 (AG1.1 to AN1) 4 100kfps N ae4(2) ae19(2) ae22(1) ae23(1)

25.0kpps 25.0kpps 25.0kpps 25.0kpps

L3VPN: ACX7509 (Unit Test) 4 50kfps N et-1/0/8 et-1/0/10 et-6/0/8 et-6/0/13

incrementing src/dst ip x3 13kpps 12kpps 12kpps 13kpps

[1]
For complete ECMP results with all outputs, contact your account representative.
[2]
Only Known Unicast is shown. VPLS BUM traffic should not load balance over ECMP. This is an expected behavior.

Convergence Briefing
Network convergences were validated against the following failure events:

• DUT Adjacent link failures

• DUT Indirect link failures
• DUT Node failures
• Daemon failure and configuration add/delete events
Overall convergence results are within expectations for the given network design. In the Fronthaul (CSR to HSR), failure
and restoration events were well within 50ms for all VPN traffic types. The latest generation ACX7000 product family
demonstrated optimal convergence, improving upon previous generation Broadcom-based platforms. For all segment
routing links, TI-LFA with loose mode allows transitioning to link-protection should node-protection become unavailable.

L2-L3 interworking scenarios with longer convergence can be attributed to control plane triggered global repair events,
which may be improved with additional mechanisms such as BGP Prefix Independent Convergence (BGP-PIC). These
models include L3VPN to EVPN-ELAN IRB Anycast Gateway, and L3VPN to Bridge Domain with IRB Static Virtual-MAC.
EVPN convergence may be further improved with dynamic-list-next-hop (DLNH) and evpn-egress-link-protection (EPL).

Table 5 and Table 6 show the convergence results in the given scenarios:

• TI-LFA without ECMP (default for the ACX7000 Fronthaul segment).

• TI-LFA with ECMP (default for MX/PTX paths in core).
• ACX7000 does not simultaneously support FRR and ECMP. During certain path failure events, TI-LFA is
prioritized over ECMP to avoid triggering global repair.

Table 5: Convergence Times for 5G Fronthaul Failure Events Per Flow Type
5G EVPN-VPWS 5G EVPN-FXC 5G L3VPN 5G L3VPN L3VPN
Flow Type
(msec) (msec) BD IRB (msec) EVPN IRB (msec) (msec)
Events CSR->DU DU->CSR CSR->DU DU->CSR DU->SAG SAG->DU DU->SAG SAG->DU CSR<>HSR

AN3 to AG1.2
7 8 0 6 - - - - 6
disable

AN3 to AG1.2
1 0 0 0 - - - - 0
enable

AN3 to AG1.1
5 1 0 0 - - - - 4
disable

AN3 to AG1.1
10 0 0 0 - - - - 1
enable

AG1.2 to AG1.1
5 0 0 0 0 0 5 0 1
disable

AG1.2 to AG1.1
5 0 0 1 0 0 0 0 1
enable

AG1.2 to AG2.2
- - - - 0 0 0 0 -
disable

AG1.2 to AG2.2
- - - - 1 0 0 10 -
enable

AG2.2 to AG3.2
- - - - 0 0 0 0 -
disable

AG2.2 to AG3.2
- - - - 0 0 1 0 -
enable

CR2 to SAG
- - - - 0 0 0 1 -
disable

CR2 to SAG
- - - - 1 0 0 0 -
enable
Table 6: Convergence Times for MBH and Midhaul Failure Events Per Flow Type
Flow Type L2VPN L2Circuit BGP-VPLS 4G L3VPN
(msec) (msec) (msec) (msec)
Events CSR<>SAG HSR<>SAG CSR<>SAG HSR<>SAG CSR<>SAG HSR<>SAG CSR<>SAG HSR<>SAG

AN3 to AG1.2 2 - 5 - 1 - 0 -
disable

AN3 to AG1.2 1 - 5 - 0 - 0 -
enable

AN3 to AG1.1 5 - 6 - 5 - 0 -
disable

AN3 to AG1.1 1 - 5 - 5 0 0 -
enable

AG1.2 to AG1.1 2 2 0 1 2 0 0 4
disable

AG1.2 to AG1.1 2 0 0 5 0 0 0 0
enable

AG1.2 to AG2.2 0 144 4 5 0 4 4 4

disable

AG1.2 to AG2.2 0 1 35 2 0 5 0 0
enable

AG2.2 to AG3.2 0 65 0 80 0 65 0 78
disable

AG2.2 to AG3.2 2 0 55 0 0 2 1 0
enable

CR2 to SAG 0 48 0 60 0 50 0 44
disable

CR2 to SAG 0 0 0 0 0 0 0 0
enable

Recommendations
The ACX7100-32C/48L and ACX7509 are excellent choices for access and aggregation purposes. They offer enhanced
performance and a wide range of advanced features, delivering significant scale and performance over previous generation
ACX platforms. Both the ACX7100 and ACX7509 are particularly well-suited for the HSR or Lean Edge metro segments, as
they provide the necessary scale, bandwidth, and performance characteristics. The ACX7100 feature-rich capabilities
surpass most access nodes requirements but are ideal for supporting 400G Fronthaul or Metro Access deployments.

EVPN-VPWS is primarily tested at scale, but Flexible Cross Connect (FXC) may further improve HSR resiliency by
aggregating services and reducing control plane scale. However, no issues were seen on ACX7000 series handling these
services at scale with E-OAM for single-homed EVPN.
Segment routing is a recommended underlay architecture to support end-to-end Seamless MPLS stitching with BGP-LU
across multiple IGP and inter-AS domains; this can be further enhanced with seamless segment routing and BGP-CT once
supported, starting with Junos OS 23.1R1-EVO. By utilizing TI-LFA and ECMP mechanisms, we can achieve fast failover
and resilience in the reference architecture. The ACX7100 and ACX7509 platforms can be confidentially deployed to
support the featured protocols and services.

While the reference design of this JVD is the 5G xHaul infrastructure, the technologies and practical solutions covered can
be leveraged as building blocks from which additional designs may evolve to support multidimensional network
architectures.

jvd-5g-fh-02-01 Revision 1.0