Virtualizing mobile networks

The silver bullet for operators to master 5G?

September 2020
Content

Executive summary 3

1. Why are new mobile architectures needed and why now? 4

2. How can virtualization be executed across network domains? 7

Conclusions 11

Authors:

Karim Taga Bela Virag

Partner Partner
TIME*, Vienna TIME, Vienna
[email protected] [email protected]

Christoph Uferer Ventsislav Dimitrov

Principal Consultant
TIME, Vienna TIME, Vienna
[email protected] [email protected]

Acknowledgement for their support and valuable input: Rui Frazao and Dominic Sattler
*TIME - Telecommunications, Information, Media and Electronics
Executive summary

Around the globe, achieving today’s customer demands of super-high throughput

and ultra-low latencies requires rethinking traditional mobile network architectures.
Monolithic telecom infrastructure based on proprietary hardware and closed
interfaces has failed to provide the flexibility, scalability and degree of automation
that is required in the era of 5G.

Technology developments in the fields of hardware and software disaggregation,

network function virtualization and containerization allow for radical architectural
changes across mobile network domains. Yet, to fully reap the benefits of network
virtualization, telco operators need to rethink sourcing and deployment models as
well as corresponding organizational setups.

This report is the first in a series of publications targeted at addressing the issues
telco organizations face in deploying their virtualized mobile networks. Here, we
focus on the general concept of virtualization and its application to the mobile
network domains. In subsequent publications, we will tackle the strategic
questions of sourcing and organizational models related to these deployments.

3
1. Why are new mobile architectures
needed and why now?

Today’s mobile networks are in a constant race to keep up metal” (i.e., running VNFs on a significantly reduced stack) to
with growing demand in coverage, capacity and customer further increase efficiency and hardware usage and to drive
experience (CEX), while average revenues per user are shrinking scalability of the network (see Figure 1).
(or stagnating in the best case). In the era of 5G, these demands Figure 1: Overview of network function virtualization concepts
are summarized in clearly defined requirements for use cases in
ultra-reliable low-latency communication (URLLC) and massive Virtual machines Container on VM Container on bare metal

data flow (enhanced mobile broadband [eMBB] and massive

(libs/VNF)

(libs/VNF)

(libs/VNF)
Container

Container

Container
VNF1 VNF2 VNF3
machine-type communication [mMTC]) connections. Demands
are further magnified by the emergence of new network Binaries/ Binaries/ Binaries/

(libs/VNF)

(libs/VNF)

(libs/VNF)
Container

Container

Container
libs libs libs Container engine
services such as end-to-end slicing and mobile private networks. Guest OS Guest OS Guest OS Guest OS

Hypervisor Hypervisor Container engine

Achieving these objectives requires rethinking traditional mobile Host operating system Host operating system Host operating system
network architecture. Monolithic telecom infrastructure based Computing infrastructure Computing infrastructure Computing infrastructure
on proprietary hardware and closed interfaces does not provide
the flexibility, scalability and degree of automation that mobile
players increasingly need. Hence, novel (for the telecom sector, Source: Arthur D. Little analysis

yet not ICT as a whole) architecture concepts are taking hold.

These include: Although some of these trends have been around for some
time, we now see three key market trends acting as catalysts
n Edge and far-edge data center (DC) infrastructure –
for the large-scale adoption of these technologies in the near
Deployment of virtual network functions (VNFs) close to the
future:
customer reduces latency, improving quality of experience
(QoE); enables higher security (especially for mission-/safety- 1. Interoperability of vendors within and between the individual
critical applications); and provides local data termination to network domains (RAN, transport, core and orchestration) is
relieve backbone network traffic. developing at a rapid pace based on standardizations such as
Open RAN (O-RAN Alliance).
n Disaggregation of hardware (HW) and software (SW) –
Unbundling of HW and SW (i.e., using commercial off- 2. As a consequence of this expansion, the vendor landscape
the-shelf compute storage and networking hardware) is has drastically broadened, leading to the availability of
nothing new, yet it only recently gained general exposure competitive solutions for all network domains on the market,
across networking equipment (especially in radio access thereby fostering a clear paradigm shift of traditional as well
networks [RANs]) to enable CAPEX reduction and sourcing as “new” network equipment providers (NEPs) to move
diversification. from monolithic telecom network infrastructure to software-
based IT solutions.
n “Softwarization” and virtualization of network functions –
End-to-end software-defined networks (SDNs) allow for the 3. We consider integration efforts and risk for decoupled
scalability and automation required for future 5G use cases. and disaggregated solutions to be reasonable given
expected network TCO savings of up to 40 percent and
n Open interfaces – Standardization of application
the establishment of an ecosystem of suitable integration
programming interfaces (APIs) between, as well as within,
partners.
network domains allows for vendor competition, hence
innovation at lower total cost of ownership (TCO).
Leading operators across the globe have already taken the first
n Containerization of software – Moving from dedicated steps toward these target networks, as illustrated in Figure 2.
servers and virtual machines (VMs) to containers on “bare

4
Figure 2: Global overview of virtual RAN/Open RAN deployments, from 2018

EXAMPLES
Vodafone currently O2 UK (Telefonica)
tendering oRAN in 14 piloting oRAN for 5G T-Mobile Poland Fully virtualized/open
European markets in rural areas planning to deploy an RAN and Core by
OpenStack core Rakuten (new entrant)

AT&T and Verizon

advanced on their own
telco cloud platforms

Multi-vendor oRAN
deployment on 5G by
NTT Docomo

Telefonica with
commercial oRAN Preparations for open
deployment in Peru oRAN by MTN (up to 4G) vRAN on 5G by Etisalat
in 21 African markets; in UAE
live in 3

Source: Arthur D. Little analysis

Although virtualization concepts and subsequent target network new 5G services such as URLLC or mMTC can be reduced from
architectures remain at an early stage of development, the first the current requirement of several days to just a few hours.
live deployments – such as those by Rakuten in Japan, AT&T
in the US or Telefonica in selected markets – clearly indicate Moreover, virtualization and cloudification allow organizations
performance improvements over distributed/centralized network to provide better QoE through continuous integration (CI)/
architectures across domains. continuous delivery (CD) and fully automated self-provisioning
and self-optimizing network (SON) functions. Figure 3 compares 2

As an example of these improvements, time to market (TTM) for key performance metrics between distributed, centralized and
incremental network deployments (mobile sites) as well as for virtualized network deployment models along the customer
journey.

Figure 3: Performance comparison of network deployment models

Customer benefit Domain/KPI Distributed Centralized Virtualized

Network deployment/incremental cell sites Weeks Days Hours

Offer best-of-
Be attracted

breed solutions
Highest quality of experience through CI/CD No No Yes
TTM

Enable new Deployment of additional services

Weeks Days Minutes
services (URLLC, mMTC)

Effective and secure 5G end-to-end

No No Yes
network slicing
Optimize service
provisioning On-demand/self-provisioning through
No No Yes
Join

process automated orchestration

Enabling SON capabilities No No Yes

CEX

DC capacity upgrade time Weeks Days Real-time

Ensure network
Use

reliability
Automated service assurance to reroute or spin
No No Yes
up resources

Source: Publicly available data; Arthur D. Little analysis

5
With a broadening vendor landscape, higher interoperability Whereas CAPEX can be reduced by up to 50 percent due to
and performance improvements along the customer journey, vendor competition across domains, OPEX savings of up to
operators such as Rakuten, China Mobile and T-Mobile US 53 percent mainly come from efficiencies in deployment and
that deploy virtual RAN (vRAN)/Open RAN (oRAN) solutions operations, such as zero-touch automation (see Figure 4).
realize network TCO savings of up to 44 percent compared to
traditional distributed/centralized RAN setups (D-RAN/C-RAN).

Figure 4: Illustrative TCO savings of vRAN/oRAN compared to traditional RAN setups

Savings potential of network virtualization Benchmarks

CAPEX savings OPEX savings Achieved savings

30-50% Mainly based on site 40-53%

rental and cooling 40%
cost reductions Claims to have 40% savings on OPEX & 30% savings in CAPEX OPEX
with open vRAN
30%
CAPEX

D-RAN/ vRAN/ D-RAN/ vRAN/

C-RAN oRAN C-RAN oRAN
53%
Reduced OPEX by 53% and CAPEX by 30% with savings in OPEX
Savings due to power consumption (cooling), site rental fees, onsite
TCO savings 30%
larger choice of management and repairs
vendors CAPEX
37-44%

Reports that oRAN reduces CAPEX for 5G by up to 50% 50%

compared to 4G CAPEX
D-RAN/ vRAN/
C-RAN oRAN

Source: Publicly available data; Arthur D. Little analysis

6
2. How can virtualization be executed
across network domains?

To realize the benefits of virtualization, organizations must revisit operational effort to maintain equipment-heavy cell sites, as
mobile network architectures across domains and explore the opposed to fewer, centralized locations.
organizational changes and alterations of traditional sourcing
models that are required. Next, we outline the key concepts for RAN disaggregation for improved efficiencies and CEX
main mobile network sections. BBUs are shifting from a physical toward a virtual resource
and can be further split into virtual distributed units (vDUs) and
Radio access network – achieving openness through virtual centralized units (vCUs). Virtual BBU (vBBU) resources
virtualization can ultimately be more efficiently allocated, increasing average
equipment utilization as a lever to reduce necessary CAPEX and
RAN assets are typically the largest investment for the mobile
improving CEX by reducing the risk of congestion. Virtualization
network and are the most rigid and costly to operate and
also enables scalability, automaticity and faster adoption of
transform. Thus, it is vital that they are made increasingly agile
changes (updates or vendor swaps).
and more cost-efficient. The evolution we see in RAN includes
centralization, virtualization and openness (see Figure 5):
Openness for vendor diversification
1. Centralization and virtualization for leaner cell sites.
RAN virtualization facilitates the implementation of open
2. RAN disaggregation for improved efficiencies and CEX. interfaces (oRAN), leading to use cases, such as deploying
separate vendors for active antenna units (AAUs) and vBBUs
3. Openness for vendor diversification.
(or vCUs and vDUs), using commercial off-the-shelf (COTS)
Centralization and virtualization for leaner cell sites hardware for vRAN software, or mixing different RAN vendors
within the same geographic footprint. Such a multi-vendor
Avoiding dedicated baseband units (BBUs) at each site allows setup allows organizations to leverage best-in-breed solutions
mobile operators to create centralized BBU pools, driving for each network component while lowering RAN CAPEX
leaner cell sites that are easier to deploy and maintain. This is through increased vendor competition and reduced reliance on
a key driver for lowering CAPEX by reducing individual cell- proprietary hardware.
site requirements for shelter and cooling, and the nascent

Figure 5: Overview of modern virtualized (vRAN/oRAN)

Centralization & virtualization RAN disaggregation for improved

1 for leaner cell sites 2 efficiencies and customer experience (CEX)

Core network
Core VNFs
vDU RAN Vendor C vCU RAN Vendor D
RAN Vendor A

COTS HW COTS HW COTS HW

Far-edge Edge
Central DCs
DCs DCs

RAN Vendor B

3 Openness for vendor diversification

Source: Arthur D. Little analysis

7
Transport – DC hierarchy and new protocols to The distributed DC infrastructure would also be shared as the
enable virtualization basis for all necessary network intelligence coming from RAN,
core or other telco domains (e.g., content delivery networks,
Virtualization and disaggregation of RAN architecture
multi-access edge computing or fixed service).
necessitates rethinking how transport and DC infrastructure is
deployed and operated (see Figure 6). The evolution for mobile
SDN up to AAU for increased controllability and
operations includes:
automation
1. Distributed DC hierarchy to allow for RAN virtualization.
As a result of RAN virtualization, two new transport domains
2. SDN up to active antenna units (AAU) for increased emerge – fronthaul and midhaul – in addition to the regular
controllability and automation. mobile backhaul. The Common Public Radio Interface (CPRI)
protocol to connect AAUs and BBUs is ill-equipped for larger
3. Fiberization for future-ready capacity.
fronthaul distances. Instead, new eCPRI and IEEE-defined
Distributed DC hierarchy to allow for RAN virtualization protocols, running on Ethernet, are necessary. To ensure
competitive quality of service despite growing user traffic and
RAN centralization and virtualization entails cloud infrastructure end-to-end orchestration, operators – and vendors – need to
at the edge. This opens an opportunity for companies to offer extend SDN controllability and automation to fronthaul transport.
latency-sensitive use cases, such as autonomous driving,
interactive augmented reality/virtual reality and holographic Fiberization for future-ready capacity
communications. To achieve this, hundreds, if not thousands, of
The traffic and latency requirements of current and future 5G
mini-DCs must be deployed closer to the user. While the build-
use cases demand a rapid increase in fiberization to cell sites.
out effort and change in operating model of the DC infrastructure
While fiber is a heavy investment, it ensures long-term capacity
may seem significant at first, the effort is a fundamental step
of the transport medium (unlike microwave, which requires
to unlocking large TCO savings from open and virtualized RAN,
frequent upgrades). Operators can approach fiber investment
as well as to increasing differentiability with better QoE and
in an intelligent way by decreasing fiber usage per site, either
available services.
at the level of transport equipment (WDM or L2/L3 switches)
As an example, whereas in a classical D-RAN concept each or at an architectural level via “ring-structure” of the fronthaul
cell site has a dedicated BBU, in a centralized and virtualized transport as opposed to the typical “star”-shaped structure of
network multiple BBUs are consolidated into centralized current backhaul. This approach will significantly reduce upfront
far-edge DCs and hosted as VNFs, eliminating significant investment into backhaul transport, while ensuring enough
amounts of remote equipment, located on the site, and thereby capacity to deliver user traffic demand and quality of experience.
reducing the overall equipment required, as well as associated With an average of around 35 percent of TCO across most
maintenance efforts. recent 5G deployments, fiberization of cell sites will remain a
key investment driver for any operator.

Figure 6: An overview of the transport architecture

1 Distributed DC hierarchy to allow for RAN virtualization

Platforms &
AAU vDU vCU
3 3 3 Peering
Fronthaul Midhaul Backhaul
3 Fiberization for future-ready capacity
Ethernet IP/Ethernet IP/Ethernet

Far-Edge DC Edge DC Central DC

2 SDN up to AAU for increased controllability and automation

Source: Arthur D. Little analysis

8
Core - distributed VNFs for critical 5G use cases provided to a specific customer. Automation will allow the
customer to directly initiate these services, enabling self-
The core network is increasingly becoming disaggregated,
service, mass customization and faster provisioning.
virtualized and distributed, enabling a more agile, efficient and
performant network that enables 5G-promised use cases. This is
mainly driven by three key factors: Orchestration and automation – moving toward
“zero-touch” operations
1. Microservice-based architecture for critical 5G use cases.
Last but not least, to achieve a design that meets operators’
2. Containers on “bare metal” for further efficiency.
business objectives, the target mobile network architecture
3. Transition to distributed core systems. requires advancements in automation and end-to-end
orchestration. Specifically, we see the following three key trends
Microservice-based architecture for critical 5G use driving end-to-end automation and orchestration (see Figure 7):
cases
1. Integration of separate domains to a single orchestration
Especially in the context of URLLC and mMTC, organizations system.
must tailor services to specific use cases. To enable these
2. From rule-based to artificial intelligence (AI)-based
differentiated services, network functions in the packet core
orchestration.
will be decoupled into individual microservices. Operators can
enhance the quality of service by catering to the demand of 3. Self-provisioning of services for full automation.
each use case, while at the same time optimizing the usage of
network resources. Integration of separate domains to a single
orchestration system
Containers on “bare metal” for further efficiency
Current network operations are based on configuration and
Moving toward a microservices-based architecture requires monitoring of separate systems – or “silos” – that must
more scalability and simplification in the orchestration of the be patched together to achieve automation and end-to-end
packet core. Hence, an evolution toward containers on “bare service provisioning. Hence, future networks will move from
metal” will be an inevitable step. Containerization allows this isolated approach toward achieving true orchestration.
core VNFs to scale-in faster and in a more cost-efficient way The journey requires initially combining various systems into
(compared to VMs) by significantly reducing the required “domains” (e.g., mobile and transport) and eventually achieving
underlaying physical resources and time. cross-domain orchestration via open APIs. This cross-domain
orchestration will enable not only a path to true automation,
Transition to distributed core systems but also the necessary capabilities for complex and dynamic
The combination of two trends – microservices and services, such as network slicing, while maintaining the
containerization – will allow operators to move from a centralized necessary quality of service across heterogenous underlying
toward a distributed core network. The latter means dynamic network systems.
and real-time deployment of microservices anywhere in the
network, based on the distinct requirements of a service

Figure 7: End-to-end orchestration architecture for network automation

E2E Orchestration and Automation

RAN Transport Core Data center …

On-demand core
functions
On-demand
Core VNFs Cloud-native,
Core VNFs
AAUs vDU vCU Core
COTS HW COTS HW COTS HW

Cell site Far-edge DCs Edge DCs Central DCs

Source: Arthur D. Little analysis

9
From rule-based to AI-based orchestration Yet along with the emergence of network function virtualization,
preventive security measures are improving as well. Concepts
Current automation is typically achieved via pre-established
such as trusted platform encryption between HW and vBBU
templates of configurations or, in the best case, rule-/policy-
and tamperproof hardware design for vBBUs have significantly
based automation. These types of automation often face
improved security levels in the RAN domain. In addition, the
a mismatch between the goals and results of the rule or
extension of IPSec and MACSec to edge and far-edge data
template, requiring additional manual adjustments. Closed-loop
center locations has provided further protection to the transport
automation would feedback from the results of actions and self-
network. Tenant isolation and full network transparency, among
adjust to achieve a target result (e.g., service quality) rather than
others, elevate security levels in the core and systems domains
a target action.
(see Figure 8).
Telcos’ end goal is “zero-touch” operations, enabling significant
OPEX reductions and better real-time optimization of network Figure 8: Overview of the security architecture

performance. AI is a necessary tool to achieve the degree of

New security measures
automation that is required. Presumably, only AI-supported
 Redundant vBBU and vBBU pooling – greater network
systems would be able to navigate the complex relations of resilience to congestion and attacks
network components. Without AI, it is unlikely that organizations RAN
 Trusted platform encryption between HW and vBBU
can achieve the dynamic service orchestration and scaling at the  Tamperproof hardware design for vBBU HW

network’s edge or network slicing.

 Distributed cloud for higher redundancy
Transport
Self-provisioning of services for full automation  Extension of IPSec and MACSec to edge/far-edge DCs

With this level of automation, our understanding of connectivity

 Simulated/testing environments
provisioning will shift to service-centric orchestration, which
 VNF lifecycle management (secure onboarding &
Core
allows for automated provisioning of on-demand services decommissioning)
triggered and configured directly by the customer. Self-  Tenant isolation

configuration enables a whole new playing field for operators

 Full network transparency
around differentiability, tapping into new customer segments Systems  Increased frequency of security updates
and further optimization potential across other commercial or  Automation and proactive threat measures
technical company units. Customer-facing self-service portals,
enabling customer-defined service parameters at low production Source: Arthur D. Little analysis

cost and fast deployment, will be a new differentiator enabled

for early adopters and a basic expectation in the long run.

Network security – increased protection through

trusted platforms and tenant isolation
Virtualization fundamentally changes the risk profile of mobile
networks. Whereas in physical networks the largest security
threat came from an “arms race” of malware with increasingly
professionalized “toolkits,” virtualization exposes network
functions to IT security threats that have not been present
before. Moreover, the usage of COTS hardware prevents vendor-
level security encryption (a trusted execution environment), and
multi-tenant use of physical infrastructure (i.e., slicing, third-party
apps) creates risks for security “backdoors” between different
security grade systems.

10
Conclusions

With changing customer requirements and increasing 2. Mobile operators will no longer be able to work in
competition, now is the time for telcos to transform their separate silos (e.g., with isolated teams for RAN,
mobile networks. Telecom operators should make use of new transport, core and systems). Rather, organizations will
technologies to transition from legacy network architecture to have to establish a “network architecture” team responsible
a flexible, virtual, open and automated network that is more for the orchestration of the entire network and to increase
resilient and programmable, enabling adjustment in the face of cross-segment coordination. Moreover, telco operators will
unforeseen future market demands. need to broaden the technical capabilities of their network
teams. These include those competencies associated with
Time will tell if telcos can achieve such a transition while IT in order to orchestrate the cloudification of all network
managing the migration effort efficiently. It is our strong segments and establish new, iterative cycles of working as
belief that such a shift requires not only significant financial opposed to rigid plan-build-run cycles.
and operational investment but also a transformation in the
organizational setup and capabilities. Telcos will no longer be In our 2019 report “Who Dares Wins!” Arthur D. Little covered
able to source and operate in their usual mode of plan – build – how CxOs, as well as operational and commercial teams, can
run. In particular, this transition will have two major impacts on address some of the challenges of virtualization. In subsequent
organizations: articles in this series, we will explore these implications in more
detail and share best practices of mobile network operators
1. Telco operators must rethink their sourcing strategy.
from around the globe that have embarked on this journey. Stay
Disaggregation of hardware and software in combination
tuned!
with subsequent network function virtualization allows for
moving toward greater vendor diversity and ultimately raises
the strategic question of whether to buy pre-integrated
solutions from traditional network equipment providers
or to source highly specialized IT-based solutions from
upcoming “new kids on the block” vendors. To answer this
question, operators must weigh cost and risk profiles versus
openness, agility and innovativeness of their target network
designs.

11
Contacts
If you would like more information or to arrange an informal discussion on the issues raised here and
how they affect your business, please contact:

Austria Japan Singapore

Karim Taga Shinichi Akayama Tomasz Izydorczyk
[email protected] [email protected] [email protected]

Belgium Korea Spain

Gregory Pankert Kevin Lee Jesus Portal
[email protected] [email protected] [email protected]

China Latin America Sweden

Yusuke Harada Guillem Casahuga Agron Lasku
[email protected] [email protected] [email protected]

Czech Republic Middle East Switzerland

Lukas Vylupek Sander Koch Michael Opitz
[email protected] [email protected] [email protected]

France The Netherlands Turkey

Julien Duvaud-Schelnast Martijn Eikelenboom Coskun Baban
[email protected] [email protected] [email protected]

Germany Norway UK
Michael Opitz Lars Thurmann-Moe Nicholas Johnson
[email protected] [email protected] [email protected]

India Poland USA

Barnik Maitra Piotr Baranowski Sean McDevitt
[email protected] [email protected] [email protected]

Italy Russian Federation

Giancarlo Agresti Alexander Ovanesov
[email protected] [email protected]
 Virtualizing mobile networks

The silver bullet for operators to master 5G?

Arthur D. Little
Arthur D. Little has been at the forefront of innovation since
1886. We are an acknowledged thought leader in linking
strategy, innovation and transformation in technology-intensive
and converging industries. We navigate our clients through
changing business ecosystems to uncover new growth
opportunities. We enable our clients to build innovation
capabilities and transform their organizations.

Our consultants have strong practical industry experience

combined with excellent knowledge of key trends and
dynamics. ADL is present in the most important business
centers around the world. We are proud to serve most of the
Fortune 1000 companies, in addition to other leading firms and
public sector organizations.

For further information please visit www.adlittle.com or

www.adl.com.

Copyright © Arthur D. Little Luxembourg S.A. 2020.

All rights reserved.

www.adl.com/MobileNetworkArchitecture