Report Scalaibility Blockchain

BLOCKCHAIN
SCAL ABILIT Y
Thematic Report Scalability, interoperability and sustainability
FOR
INTEROPERABILIT Y
GOVERNMENT
AND
AND PUBLIC
SUSTAINABILIT Y
SERVICES
OF
BLOCKCHAINS
a
a thematic
thematic report
report prepared
prepared by
by
THE
THE EUROPEAN
EUROPEAN UNION
UNION BLOCKCHAIN
BLOCKCHAIN
OBSERVATORY
OBSERVATORY & FORUM
AND FORUM
An initiative of the
1 An initiative of the
About this report

The European Union Blockchain Observatory & Forum has set as one of its objectives
the analysis of and reporting on a wide range of important blockchain themes, driven
by the priorities of the European Commission and based on input from its Working
Groups and other stakeholders. As part of this it will publish a series of thematic
reports on selected blockchain-related topics. The objective of these thematic reports
is to provide a concise, easily readable overview and exploration of each theme
suitable for the general public. The input of a number of different stakeholders and
sources is considered for each report. For this paper, these include:
• Members of the Observatory & Forum’s Working Groups.

• ‘Blockchain scalability, interoperability and sustainability’ by Arthur Gervais – an
academic research paper prepared by the Lucerne University of Applied Sciences
and Arts – an academic partner of the EU Blockchain Observatory & Forum.
Available online at www.eublockchainforum.eu/reports.
• Input from participants at the ‘Scalability, interoperability and sustainability’
workshop held in Berlin on 2 October 2018.
• Input from the Secretariat of the EU Blockchain Observatory & Forum (which
includes members of the DG CONNECT of the European Commission and
members of ConsenSys).
CREDITS DISCLAIMER
This report has been produced by ConsenSys The information and views set out in this
AG on behalf of the European Union publication are those of the author(s) and do
Blockchain Observatory & Forum. not necessarily reflect the official opinion of
the European Commission. The Commission
Written by: Tom Lyons, Ludovic Courcelas, does not guarantee the accuracy of the data
Ken Timsit included in this study. Neither the Commission
Thematic Report Series Editor: Tom Lyons nor any person acting on the Commission’s
Workshop moderator: Reto Peter Gadient behalf may be held responsible for the use
Report design: Benjamin Calméjane which may be made of the information
contained therein.
v1.0 - Published on 6 March 2019.
2
ACKNOWLEDGEMENTS NOTE
The authors would like to expressly While we have done our best to incorporate the
acknowledge the following for their direct comments and suggestions of our contributors
contributions and feedback to this paper: where appropriate and feasible, all mistakes
and omissions are the sole responsibility of the
Observatory Working Group Members: authors of this paper.
• Javier Sebastian Cermeño
• Tamás Chlepkó
• Cristina Carrascosa Cobos
• Arnaud Le Hors
• Phillip Sandner
• David Suomalainen
• Konstantinos Votis
• Vlad Zamfir
Scalability Workshop Panelists:

• Peter Broadhurst
• Cyril Cassagnes
• Jesus Ruiz
• Gilbert Verdian
3
Contents
5 Executive summary
7 Introduction: The blockchain landscape in Europe

A multiverse of chains
A global “backbone” of decentralised chains

7
9
10 Main issues and challenges in blockchain

Scalability
Interoperability
10
11
Sustainability 12
14 Success factors for large-scale blockchain projects
Clear purpose and concept 14

Governance 14
Component-based, service-oriented architecture 15
Homogeneous production environment and pooled resources 15
17 Considerations for a European blockchain

infrastructure
Goals of a public services blockchain infrastructure 17
Key principles 17
The layers of a blockchain infrastructure 18
20
Recommendations
22 Appendix
4
Executive summary
In this paper we take a look at the current and likely future state
of blockchain in Europe through the lens of large-scale blockchain
platforms. Along the way we ask ourselves what factors, technical and
organisational, are likely to shape how platforms develop, and make
some observations and recommendations for entrepreneurs and policy
makers about best practice.
The timing seems right for such a paper. The blockchain ecosystem
has been steadily maturing over the past several years, and projects are
getting both larger and closer to going live, or have already dones so. This
is exciting, as we can expect several large platforms to attract significant
user bases over the course of the year. Observing these projects and the
development of the technology also allows us, perhaps for the first time,
to see the outlines of what a “live” blockchain ecosystem might look like
in the near term, as well as identify the main challenges and success
factors involved.
Our vision is that the first wave of blockchain adoption will be

characterised by a large number of permissioned, purpose-built
blockchain platforms geared towards a specific use case or user base.
These blockchains will, however, not be completely walled off gardens.
Instead, they will need to interact with the off-chain world as well as with
each other. Just as TCP/IP and the rest of the Internet stack became the
open, freely accessible backbone of the Web of Information, we think
that a small number of global blockchain networks will also emerge as
the backbone of a Web of Value. To get there, the blockchain community
will need to solve an array of challenges that we have gathered into three
categories:1
• Scalability: the ability to handle large volumes of transactions at high

speeds.
• Interoperability: the ability to exchange data with other platforms,
including those running different types of blockchains, as well as with
the off-chain world.
• Sustainability: a) the ability to run a large-scale blockchain platform
or decentralised application in an environmentally responsible
way, and b) the ability to govern projects, platforms and the core
technology in such a way that they remain viable over the long term.
1 Note that, while they represent important challenges for the blockchain ecosystem as well, we do not explicitly address
issues of privacy and confidentiality in this paper. We intend to handle these themes as part of our workshop and thematic
report on security to be held in the second half of 2019.
5
EXECUTIVE SUMMARY
A lot of time and effort is being expended on surmounting the above

challenges, and we examine each in some detail both in the main text
and in a series of technical “deep dives” in our appendix. Privacy and
confidentiality, the last main technical discussion not mentioned so far,
will be covered extensively in another paper.
As an increasing number of large projects reach maturity, we can also

begin to identify the characteristics of the successful ones. In sum,
we believe that projects need a clear vision of what they want to
accomplish, a clear reason for using blockchain instead of traditional
database technology, and strong governance structures that provide
clarity on roles and responsibilities and support collaboration and
sharing of effort and expertise among diverse stakeholders.
We think government can play a role in fostering success as well through

supporting the development of the base blockchain infrastructure. We
therefore take a look at the key success factors for such an infrastructure,
and some of the principles we believe policy makers should follow to
support its development.
We end with a set of recommendations as well. First priority remains

basic research: Europe has been very supportive in this area, but there
is much still to be discovered and developed. To ensure its place as a
leader in this new technology, Europe will need to continue to fund work
on next-generation solutions. As blockchain matures, there will be an
increased need for standards both for the technology as well as how best
to work with it (governance). Getting this right will require a balancing act
between harmonisation and fostering technical diversity. We therefore
believe that a light-touch approach, allowing for experimentation, is
the right one for the moment. Last but not least, we believe European
governments need to be open to these changes, employing blockchain
themselves in government services where it makes sense and so
preparing themselves for the potential mass adoption of this technology.
The good news is that Europe already has a relatively good track record
in all of the above. That makes us optimistic that our vision of a maturing
ecosystem on the brink of mass adoption is a correct one.
6
Introduction: The blockchain

landscape in Europe
Survey the blockchain ecosystem in Europe today1 and you will find
an impressive number of startups and projects building platforms and
decentralised applications, not to mention literally thousands of white
papers, proofs of concept and Github repositories. You will also find a
vast and vibrant community of associations, non-profit organisations,
foundations, meetups and local, national and supranational government
initiatives working to further the development and adoption of this new
technology. What is harder to find are live platforms or applications with
large, active user bases.
We think this is about to change.
If 2016 was the “year of education”, when people learned about

blockchain and its uses, 2017 the “year of proofs-of-concept”, during
which people experimented to validate theses, and 2018 the “year of
large-scale projects”, with a number of significant projects announced
and in development, there is now very good reason to believe that 2019
will indeed be the “year when projects go live”, with a number of major
platforms slated to go or already in production.
A by-product of this evolution is a bit more clarity: we begin to see the

outlines of what the technology and ecosystem will look like in the near
future, what the hurdles to mass adoption are, and what some of the best
practices and success factors for surmounting these hurdles are likely to
be. This, broadly speaking, is the subject of the present paper.
A MULTIVERSE OF CHAINS
So where is blockchain adoption headed in the near future?
From today’s vantage point, it seems likely that the first successful
implementations of large-scale, live blockchain platforms will come
in the form of very efficient, “permissioned” blockchains developed by
1 A good place to start is the ‘EU Blockchain Map’ we have developed at the EU Blockchain Observatory & Forum, which
counts over 550 entries and is continuously growing.
7
INTRODUCTION: THE BLOCKCHAIN LANDSCAPE IN EUROPE
industry or other types of consortia, and focused on a single sector and/or

use case.2
Such consortia can take on many different forms and involve different
kinds of stakeholders. Below are examples of three of the most prevalent
consortia models:
• Multi-stakeholder-managed industry consortium. A popular

approach involves a group of market participants (who might very
well be competitors) that come together to build a market platform.
Such consortia are governed collectively by their members.3
• Single stakeholder-managed, blockchain-based industry
ecosystem. In other cases, a single provider may build a platform and
open it up for others to use.4
• Geographically based blockchain consortia. Another approach,
less common but with interesting implications, is to build a general
purpose, large-scale platform for more or less general transacting
within a defined community that provides an infrastructure for
members to build upon.5
There are many advantages to such market- or use-case-focused

approaches.
Limiting the blockchain to a single or small set of related use cases

means that developers can design the platform to meet the specific
needs of the users. This provides a great deal of flexibility. For example,
builders of “private” blockchains currently have more leeway to design
for performance and security than designers of public blockchains do
(more on this in our discussion of the blockchain trilemma in the next
section). A private or semi-private undertaking with a limited number of
stakeholders and a well-defined focus will also likely have an easier time
of defining and agreeing to a governance model than is the generally the
case with large, open, public platforms. Finally, considering how new, and
in certain respects untried, blockchain technology is, working in a “walled
garden” can be prudent from a software development and data security
perspective.
2 For a short overview of blockchain, see Appendix 2 below.

3 A good example is Komgo: https://komgo.io/
4 See for example the IBM Food Trust: https://www.ibm.com/blockchain/solutions/food-trust
5 The pioneer here is Alastria in Spain: https://alastria.io/index_en.html
8
INTRODUCTION: THE BLOCKCHAIN LANDSCAPE IN EUROPE
A GLOBAL “BACKBONE” OF DECENTRALISED CHAINS

There is, however, a flaw in the walled-garden approach: the walls.
Networks are powerful in proportion to their size. As we know, large

private versions of the Internet like AOL eventually disappeared as the
maturation and adoption of the world wide web provided average people
with user-friendly access to the open, decentralised, global version. This
model won the day mostly because it proved the most inclusive and most
useful.
Something similar is likely to happen in blockchain.
It seems clear to us that a multiverse of independent blockchains that

cannot interoperate would be severely limited. Users of blockchain
platforms will find it beneficial to be able to exchange data and make
transactions between chains too: a healthcare chain connecting to
an insurance chain, a real-estate chain connecting to a construction-
materials or manufacturing chain, and so on.
Just as TCP/IP and the rest of the Internet stack became the open, freely
accessible backbone of the Web of Information, we think that a small
number of global, decentralised blockchain networks will also emerge as
the backbone of a Web of Value.
These will allow these permissioned chains to work with each other by
helping to facilitate data exchange and interoperability between chains,
or even simply serving as a timestamp and settlement layer for other
chains, and so providing that key layer of trust.
This is, however, by no means the only possible scenario. We may

also see interoperability technologies arise that allow for seamless
interaction between blockchains, resulting in a mesh network of directly
interoperable chains.
9
Main issues and challenges in

blockchain
SCALABILITY That said, compared to conventional financial

transaction platforms, the Bitcoin blockchain
is incredibly slow, processing some seven
If blockchain-based platforms and applications transactions per second (tps) on average. Many
are to succeed, they will need to be able to of its blockchain peers do somewhat better,
scale sufficiently to meet the needs of their with different configurations able to achieve
target markets or users. Due to the way 10x or even 1,000x this rate. All of which,
blockchains work, scaling poses challenges however, pales against VISA’s 24,000 tps.
that are not present with conventional
database technology. With the current state of blockchain
technology, this lack of performance cannot
The point of a blockchain is to create be avoided, at least in public, permissionless
decentralised trust through distributed blockchains: it is part of the price of securing
ledgers that are verified and maintained the network.
by a community of peers. In the original
Bitcoin blockchain, this is achieved by having In the blockchain world people often refer to a
all transactions propagated throughout loose trilemma positing that blockchains can
the network and then having miner nodes generally have only two of the following three
compete to verify the transactions and append properties: scalability (that is, performance in
a block of transactions to the ledger, which in terms of speed and volume), decentralisation
turn is propagated to all the full nodes before or security.1
the process starts over.
If a blockchain is to be highly decentralised
The competition to validate the transactions and highly secure, it will come at the cost
uses what is known as a proof-of-work of scalability. If it is highly performant and
consensus mechanism, which can be highly decentralised, it will not be secure.
described as a kind of computationally Similarly, if one is willing to accept a degree
expensive lottery. This lottery forces the miners of centralisation, it is possible to build highly
to pay a price (in hardware, energy costs and secure and performant blockchains.
time) to take part. That price, in turn, secures
the network, among other things by making it As a result, designers of blockchain-based
prohibitively expensive to cheat.
platforms need to consider the trade-offs
between these three parameters that best
Most, though not all, blockchain technologies
fit their particular use case.
today follow a similar approach, and for a good
reason: it works. Despite holding billions of
1 See https://github.com/ethereum/wiki/wiki/Sharding-FAQs#this-sounds-like-
dollars of value, the Bitcoin network has never theres-some-kind-of-scalability-trilemma-at-play-what-is-this-trilemma-and-can-
been compromised. we-break-through-it
10
MAIN ISSUES AND CHALLENGES IN BLOCKCHAIN
INTEROPERABILITY
As we have mentioned, for blockchain-based
platforms to succeed they will need to be able
to communicate and share data, a property
that is usually referred to as interoperability.
While interoperability between blockchains
can be achieved in a number of different ways,
these can be broken down into two basic
categories.
One category of interoperability approaches

involves the use of trusted third-party
authorities to validate transactions or
information. Two or more blockchains, for
If large-scale decentralised trust is
example, might agree to trust an off-chain
important, as is the case for example with entity to either transfer information between
a global cryptocurrency like Bitcoin, then the blockchains (an exchange) or to record
decentralisation and security will likely be the state of the respective blockchains so that
favoured at the cost of scalability. In closed each can trust what has transpired on the
markets like those created by a consortium, other (notary services). A notary service in turn
where it is possible to control who is can be either fully centralised (a single trusted
allowed to participate on the network and third party) or federated (a group of providers
of these services). Multiple blockchains can
to “police” their behaviour through both
also rely on an outside data source to provide
on-chain and off-chain means (for example
trusted reference information, for example
legal contracts), full decentralisation is less
certificates that prove the identity of a person
important. That leaves room to design for
or asset. The common thread in all these cases
security and faster performance. is that trust is external, vested in some off-chain
authority.
Solving the scalability issue for blockchains
– and in particular public, permissionless The other category of interoperability
ones – is one of the most important approaches involves sharing information
technical challenges facing the community directly between blockchains without the need
today. There are many different approaches for a third-party authority. Such solutions often
employ other blockchains or smart contracts to
currently being researched and developed,
supply the trust needed to carry out inter-chain
and as these challenges are met, we may
transactions or exchange data, thus creating a
be able to solve the trilemma.
bridge directly between chains.
We look more closely at the technology of Like scalability, interoperability is another of the
scalability in our deep dive in the appendix. key technological challenges in the blockchain
11
world at the moment. A lot of time and There are different ways to address the
effort is being invested in developing various environmental sustainability issues of
approaches and solutions, and we can expect blockchains. The most promising is by
that over time it will become easier and easier moving to a different consensus mechanism.
for disparate blockchains to work together. And indeed, much of the research and
This will be to the benefit of the ecosystem as a development in blockchain today revolves
whole. around developing alternative consensus
mechanisms, such as proof-of-stake, that use
We also discuss interoperability in more detail significantly less energy while delivering the
in our deep dive appendix. same amount of security. We discuss a number
of these in our deep dives.
SUSTAINABILITY While these alternatives are all very promising,

and while the overall trend in the industry is
Finally, successful blockchain projects will need to move towards more sustainable consensus
to be sustainable, in the sense of being viable mechanisms, at the present moment proof-of-
over a long period of time. When we speak work remains the most widely used and battle-
of sustainability, we mean one of two quite tested form of blockchain consensus in public
different things. networks.
1. Environmental sustainability of Those employing proof-of-work also have

blockchains options to reduce their impact. They can – and
indeed often do – set up their server farms in
As is widely known, updating and securing areas with cheap, abundant or clean sources of
the Bitcoin blockchain – as well as other energy. It is also possible to re-cycle the energy,
blockchains that make use of proof-of-work for example by using the heat generated by the
consensus – requires a great deal of electricity. servers to heat buildings.
By recent estimates, Bitcoin consumes the
equivalent amount of electricity in a day as the That said, it is clear that the blockchain
country of Singapore2. community considers environmental
sustainability a priority concern, and we can
As we have already touched on, this is not a expect blockchain technology to become less
bug in the blockchain protocol, but rather energy-intensive over time.
a feature. The proof-of-work consensus
mechanism used in Bitcoin is expressly 2. The sustainability of blockchain
designed to be computationally expensive: projects, protocols and ecosystems
it forces miners competing to win mining
rewards to run their computers for a certain One of the peculiarities of blockchain
period of time, and that in turn costs money in technology is that many of its most significant
the form of the electricity needed to run the core technology efforts are open-source,
hardware. community-run projects.
2 See https://digiconomist.net/bitcoin-energy-consumption.
12
Bitcoin, for example, is a purely open-source

project with no formal governance structures,
developed and managed by a more or less
organised group of developers who themselves
are often volunteers.
Many of the other significant blockchain

protocols under development are open-source
projects that have raised funds through an ICO
or via sponsors, and are managed by a non-
profit foundation.
Such projects are often dependent upon

an ecosystem of developers and others for
contributions to their ongoing development.
For anyone relying on these technologies, the
sustainability of that ecosystem is therefore of
paramount importance.
Sustainability can be measured along a

number of different parameters: the source
and amount of funding, the quality of
the governance structures, the size and
cohesiveness of the development community,
and so on. For many people, the market
capitalisation of the token associated with the
protocol can also serve as an indicator of the
health of the project.
In the next section we analyse some of the

key success factors for large-scale blockchain
projects.
13
Success factors for large-scale

blockchain projects
As we now have a number of major projects This requires asking questions like what level
that have carried out successful proofs-of- of trust should be placed in the validating
concept and are close to going or have gone nodes, what level of performance is required
live, we can begin to identify some of the or what are the acceptable trade-offs between
common elements that lead to success. In our scalability, security and decentralisation. A
opinion, these include the following.1 small consortium of stakeholders well known
to each other, with relatively low volumes of
transactions, will have different needs in terms
CLEAR PURPOSE AND CONCEPT of scaling, for instance, than a project catering
to a broad user base.
We think the most important success factor
is not technical, but conceptual. More than
anything else, projects should be clear about GOVERNANCE
the goals of their endeavour. Without a strong
vision and purpose, projects risk failing. Whether it is building or running a large-scale
project or consortium, successful collaboration
Once the use case is clear, stakeholders should requires strong governance. This can be
also be sure of the benefits of decentralisation challenging, particularly since the question
for their particular use case. If decentralisation of governance in collaborative consortia for
of some kind does not provide obvious decentralised technologies is still relatively
advantages, then it may very well be that new, and there remains a lot to be learned.
a traditional database provides a better That said, we believe the following is important.
alternative. This is a good thing to know before
investing time and effort in a potentially The consortium will want a good set of terms
unnecessary blockchain solution. and conditions, clearly spelling out such things
as the requirements to participate in and/or
Provided it is a blockchain use case, then leave the consortium, a mapping of the on-
projects will want to identify the key chain proofs with the relevant off-chain legal
functionalities of the platform needed to and regulatory frameworks, clear service-level
address the needs of the user base and agreements spelling out each stakeholder’s
particular use case, and then choose the most rights and duties, and so on.
suitable blockchain technology for the job.
In a permissioned blockchain, which these
projects mostly are, perhaps the most
important element of governance is identity.
1 This section is based on the presentation of Peter Broadhurst at the EU Projects have to consider how they will
Blockchain Observatory & Forum workshop on Scalability held in Berlin on 2
October 2018. The video is available here: https://www.eublockchainforum. establish the identity of members and what
eu/video/workshop/scalability-interoperability-workshop-2-october-2018-ber-
the rights and responsibilities of the various
lin-part-4-presentation
14
SUCCESS FACTORS FOR LARGE-SCALE BLOCKCHAIN PROJECTS
roles on the platform are. They will need to A permissioned network will, for instance,
think of the appropriate participation levels need some kind of identity registry to hold
and how entities qualify for different roles, the credentials of its users for authentication
while having mechanisms in place to ensure purposes. Most if not all of the user data will
that all members are treated fairly. There also typically be kept off the chain, so the
should be clear criteria for who can vote on project will need a shared but encrypted and
consensus, who is allowed to have a complete access-controlled data repository. There should
copy of the chain (which may contain sensitive be a means for members of the consortium to
data), who can put data into the chain and communicate and share high volumes of data
who can process that data. with each other in a secure way. Then there
are the various connections to the outside
In a collaborative IT project, especially a big world, whether data oracles or transaction
one, participants will likely want to use shared submissions. A key management solution, with
tools and IT standards as much as possible, yet recovery options, is also something projects
these must also be compatible with the tools should consider.
and standards of each individual stakeholder.
Such tools and standards should be considered Projects should further keep in mind that
carefully and, to the extent this is feasible, blockchain and other technologies evolve
agreed upon ahead of time. quickly. It therefore makes sense to use a
component approach for the above-mentioned
Finally, projects should keep in mind that they layers, developing them as microservices that
will grow. A governance model that works for a can be swapped out as new options become
small group looking to quickly build a proof-of- available.
concept to bootstrap an idea will not work for a
large, live platform with hundreds of members.
The governance model should therefore also HOMOGENEOUS PRODUCTION
be flexible, allowing the project to evolve and
adapt as necessary. ENVIRONMENT AND POOLED
RESOURCES
COMPONENT-BASED, SERVICE- Sharing is also a key factor. In successful
ORIENTED ARCHITECTURE projects, members tend not to build their apps
to connect to the platform, but customise
In designing their solutions, projects should be standard apps developed by the consortium,
mindful of the fact that the blockchain is often saving time and effort and ensuring
only a small part of the overall architecture and compatibility. This can, however, be a challenge
hence of the development effort. There will be especially when dealing with groups of well-
many other layers that need to be understood established enterprises with a long history
and engineered to fit into the platform as a of IT projects and their own well-developed
whole. set of standards, as well as differing levels of
blockchain expertise.
15
SUCCESS FACTORS FOR LARGE-SCALE BLOCKCHAIN PROJECTS
For this reason, successful projects tend to

have a centralised, cross-funded shared IT
organisation – staffed either by individuals
from the member organisations seconded
to the project, or set up as a separate funded
entity – that develops common practices and
tools for the consortium. This organisation
can act as a shared pool of skill, resources
and expertise, playing an important role not
only in developing the platform and keeping
it running, but also in answering questions,
providing support and generally keeping the
wheels greased.
16
Considerations for a European

blockchain infrastructure
A new technology can’t become successful on of an innovative, world-class blockchain
a mass scale without the right infrastructure. ecosystem in the EU.
This includes such things as the technical • Make it easier and cheaper for public and
hardware and software stacks, standards, the public/private initiatives to design, test
legal framework as well as governance best and deploy projects based on blockchain
practices. technology.
• Establish interoperability standards that
Through the European Blockchain Partnership make it possible for individual projects to
the EU has begun to consider Member State communicate with each other easily and
cooperation towards the goal of a European securely, with no leakage of sensitive data.
Blockchain Services Infrastructure to support • Facilitate knowledge transfer between
cross-border digital public services.1 projects.
In this section, we offer some general

thoughts on some of the considerations that KEY PRINCIPLES
could influence the development of such an Such an infrastructure policy should consider
infrastructure, and draw a high-level picture of the following principles and key points:
how such an infrastructure might look.2
• Technological diversity. At this early
stage of the technology, it would make
GOALS OF A PUBLIC SERVICES sense for the EU Blockchain Infrastructure
to hedge its bets by allowing multiple
BLOCKCHAIN INFRASTRUCTURE technologies to operate and compete
on the platform. Similarly, it could be
We think that government can play a argued that, by allowing multiple identity
positive role in helping to develop such an solutions, payment solutions and other
infrastructure through policy. To foster the technology components to operate on the
Digital Single Market we believe it makes sense EU Blockchain Infrastructure, the platform
for European policy makers to focus on the will protect itself against vendor lock-in
following goals for a blockchain infrastructure and will encourage providers to develop
initiative: interoperable solutions.
• Interoperable communications. The EU
• Create an environment and a set of should foster interoperability through
foundations conducive to the development various policy means. Interoperability is a
functionality offered by a specific tool or
1 See https://ec.europa.eu/digital-single-market/en/news/european-countries- library, and each tool or library is usually
join-blockchain-partnership
2 The text for the rest of this section first appeared, in slightly different form,
customised to an origin, a destination,
in “A public blockchain infrastructure in the European Union?”, by Ken Timsit, 14 a type of data and the level of trust that
December 2018.
17
CONSIDERATIONS FOR A EUROPEAN BLOCKCHAIN INFRASTRUCTURE
the origin and the destination have in • Back-ups. The EU Blockchain Infrastructure
certain “signing authorities” associated should allow users to request regular back-
with the data. We cannot request that a up of their server instances and their data.
EU Blockchain Infrastructure should have • Roll-back / fork. The EU Blockchain
“interoperable blockchains”. Rather, we ask Infrastructure should, at some point in
for specific interoperability functionalities, time, offer “roll-back” functionalities. While
and expect these functionalities to expand blockchains are, in principle, immutable,
over time. there may be especially critical situations
• Digital identity. Blockchain-based where processes or data have been
government services applications are corrupted due to a bug or a security issue,
unlikely to be able to operate without making it necessary to roll-back or fork one
a strong underlying digital identity of the blockchains. Such roll-backs or forks
ecosystem. A decentralised identity should be exceptional, and decided only
ecosystem includes a wide universe of by the highest governance bodies of the
applications, devices and authorities, Infrastructure platform. Nevertheless, if they
allowing users to store identity credentials happen, it is important to have pre-existing
in the repository of their choice and governance, processes and functionalities
use them across multiple blockchains in place in order to make them happen
and applications. Such a decentralised quickly.
identity ecosystem could be designed to • Migration / living wills. The EU Blockchain
interoperate with the centralised national Infrastructure should not, in principle, lock
digital identity schemes and eIDAS, the the data of public services within a specific
European framework and platform for such blockchain technology. Governing bodies
schemes. should request that technology vendors
• Fiat currency payments. Many blockchain- generate “living wills”, i.e. high-level guides
based workflows involve some form outlining how the data could be migrated
of automatic payment when certain out of their technology into a standard
conditions are met (such as the completion database, should the need arise.
of a workflow). Given that blockchains and
banking systems are often two separate
environments, these workflows are unlikely THE LAYERS OF A BLOCKCHAIN
to be able to operate unless fiat currencies
are given an existence on blockchains, and INFRASTRUCTURE
can be manipulated by smart contracts. The
EU Blockchain Infrastructure could allow In order to meet the functional needs of
users to access tokenised fiat currencies, institutional users and end users, a blockchain
for example by allowing private financial infrastructure should likely consist of the
institutions to issue tokens in proportion following layers, from top to bottom:
of the amounts held by these institutions
in escrow accounts or by working together • An application and services ecosystem/
with the European Central Bank to issue layer, consisting of applications focusing
a central bank-backed form of tokenised on specific use cases, service providers,
euro. auditors and regulators, and governance
18
CONSIDERATIONS FOR A EUROPEAN BLOCKCHAIN INFRASTRUCTURE
bodies; some of these applications and

services could be made available to
end users via some form of app store or
marketplace.
• A platform presentation layer, consisting
of consoles, dashboards, and development
environments made available to
developers, institutional users, auditors and
regulators.
• A middle layer of services, callable by
the above layers via APIs by entities and
users who have the required permissions.
Such services would include core
APIs for reading/writing data onto the
various blockchains and databases, as
well as shared utilities such as identity,
interoperability bridges, event listeners
and oracles, and perhaps even a payment
infrastructure.
• A platform management layer, consisting
of a registry of networks and participating
nodes/servers and users as well as their
respective permissions, as well as various
monitoring tools.
• A blockchain and database layer,
containing the actual blockchain with the
chosen consensus mechanism as well as
any ancillary databases.
• A marketplace of templates, consisting
of a menu of images of machines and
containers that can be selected and
deployed by institutional users to create
as many servers/nodes as they see fit,
including blockchain clients, and other
technology stacks such as databases, web
servers and others.
• A cloud infrastructure, in which
institutional users can select and deploy
as much computing and storage power as
they need for their respective application.
19
Recommendations
Our purpose in writing this paper has been to in supporting blockchain research in a
shed some light on the “potential future states” number of ways, from the EU Blockchain
of blockchain infrastructure and platforms in Observatory & Forum, under whose
Europe assuming large-scale adoption, and aegis this paper has been written, to the
to examine the characteristics for success in allocation of up to EUR 340 million to
getting there, as far as they can be ascertained support blockchain projects through 2020
at this time. under the Horizon 2020 programme. While
these are laudable, the EU should not rest
As we hope we have been able to show, it on its laurels. Both the US and China have
is indeed an interesting moment for the expressed strong support for blockchain
blockchain ecosystem and industry in Europe, research, with the former even going so
as a number of large projects are beginning far as to include it as part of its USD 700
to go live and attracting large user bases. billion defence budget. We therefore
While this is encouraging, there are still many recommend that the EU continue its strong
challenges – both technical and organisational support, targeting both basic research
– that need to be surmounted. as well as supporting implementation of
infrastructure-related projects in particular,
We believe there is a role for policy makers to as well as research into non-technical
help support the industry in meeting these topics such as governance of blockchain
challenges, and the good news is that the EU projects (see next bullet).
has already shown a clear intention to do so. • Governance. Governance, as we’ve
In this last section, we offer a few concrete mentioned, is a very important topic and a
recommendations for policy makers to help key success factor for blockchain projects.
continue the momentum. Governance of decentralised technologies
and applications is also not very well
• Standards. Today if we want to go in the understood. Besides supporting research
direction of multi-layered interoperable into the topic, the EU should take a wait-
ecosystem of blockchains, both policy and-see approach, giving projects the time
makers and the blockchain industry, to experiment and learn before developing
ideally working together, should prioritise standards or considering governance-
the development of standards. Among related regulations.
these, we think the most important will • Ecosystem diversity. The EU can support
be standards for digital identities in a the ecosystem in other ways as well, for
blockchain context, and for interoperability instance by encouraging project diversity.
between blockchains. We think it best, for example, not to
• Research. It is no secret that research mandate specific technologies under the
is an important element to the success European Blockchain Infrastructure, but
of blockchain. Currently the EU is active rather encourage different technologies
20
RECOMMENDATIONS
to foster innovation. Policy makers should

encourage fiat money on-chain to facilitate
blockchain-based payments and the
uptake of smart contracts. An open data
policy within government, providing
bridges to publicly available information
in government databases to blockchain-
based systems, would help foster the
technology, .
• Be blockchain-aware themselves. On a
broader level, policy makers will want to
be mindful of and prepared for the new
models and possibilities being brought
about by decentralised technologies. As
well as using blockchain themselves to
provide government services,1 governments
could well be interacting with blockchains
as identity providers or data oracles. They
will want to be prepared.
• Legal Framework. As we have mentioned
in previous papers and workshops, the
success of the blockchain industry in
Europe will depend to a large extent on
clarifying many of the legal and regulatory
issues thrown up by blockchain. These
include resolving the tensions between
GDPR and blockchain, the legal, fiscal and
accounting status of crypto assets, and
the legal status of smart contracts, among
others.
• Education. Along with research, blockchain
adoption will also depend on the
education and training of technologists,
entrepreneurs, other experts and the
general public. This is an area where policy
makers can have a significant impact
through the support of educational and
research initiatives.
1 See Blockchain for Government and Public Services, EU Blockchain Observa-

tory & Forum, 7 December, 2018.
21
Appendix — Deep Dives
BLOCKCHAIN LAYERS
When dealing with scalability, there are multiple technologies that
one can use or improve in order to make a solution work faster. These
technologies are related to four different layers:
Layer -1 solutions: Hardware Layer

The hardware layer refers to the machines that are used to run the
blockchain client (software). Using better machines can translate into
better performance in terms of transaction throughput as the basic
computations (e.g. verifications, changes of state) can be performed
more quickly. It is important to highlight that all machines on the
network need to be upgraded in order to actually realise performance
improvements. This solution is often leveraged for specific cases such as
consortium chains and dPoS, where a small number of actors are able to
deploy high-end hardware.
How much scalability can you gain through this layer?

Up to 5 to 10 times more throughput.
Limitations
This solution cannot be easily leveraged for large networks because
machines are heterogeneous performance-wise, meaning that the older
nodes would not be able to keep up if the settings make it too hard to
keep pace. Minimum requirements could be contemplated for machines
to run a client but it means that only a limited number of actors would
be able to afford the necessary hardware, resulting in a less decentralised
system.
Layer 0 solutions: Network Layer

Layer 0 refers to the network layer, including all communications
between nodes. An important concept to keep in mind is the one of
network propagation, which describes how fast a certain percentage
of the network will have received the latest bits of information (e.g. the
last block). There are physical limitations to how fast information can
travel around the globe, but also core parameters that will influence
22
APPENDIX — DEEP DIVES
propagation such as what data is being sent, how much there is of it

and which transmission method is being employed. Most blockchain
protocols rely on peer-to-peer models for communication between
nodes.

Up to 5 times more throughput.
Layer 1 solutions: Blockchain Layer

There is a lot of work going on in the development community to improve
the performance of blockchains themselves, also referred to as Layer
1. Innovating on protocols can lead to systems that are faster or more
efficient (i.e. use less energy).
Layer 1 solutions can be minor tweaks, for example adjusting the block
size or the block time interval. Or they can be major changes. Blockchains
can employ sharding, for example, which is a technique common in
conventional database technology. This is where you break down the
data to be processed into different partitions (shards) and have part of
the network validate the individual partitions, thus allowing for parallel
processing.
Another approach is to change to a different consensus mechanism. The

proof-of-work consensus mechanism is deliberately slow and costly: that
is its main means of maintaining the security and viability of the network.
There are, however, many other possible consensus mechanisms. The
most promising at the moment for public networks seems to be Proof
of Stake, which itself comes in a number of variants. There are also non-
blockchain systems that allow for decentralised trust through distributed
ledgers, such as the Directed Acyclic Graph (DAG). It is, however,
important to keep in mind that different consensus mechanisms imply
different trust and security assumptions, among other things, and so
cannot be compared to each other simply on the basis of transactions
per second or other performance metrics.
Focus on consensus algorithms:

The choice of a consensus algorithm fundamentally affects the
characteristics of a blockchain, including the stability of the network,
scalability, energy consumption and resilience to failures and malicious
attacks. Each consensus algorithm has its pros and cons, and some
will be preferred in specific contexts. Even though there are dozens of
different types of consensus algorithms, each with multiple different
23
implementations, some of them are most often used.
Category of Used for Time before # of nodes on Tolerance to Energy

consensus finality the network malicious consumption
algorithm participants
Proof of Work Public High High 1/2 High
networks
Proof of Stake Public Low High 1/3 Low
networks
Delegated Public Low Low 1/3 Low
Proof of Stake networks
Practical Permissioned Low Low 1/3 Low
Byzantine networks
Fault
Tolerance
RAFT/PAXOS Permissioned Low Low 1/3 Low
networks
Fig. 1 Comparison of some consensus algorithms

Up to 10 to 20 times more throughput.
Limitations
It is important to keep in mind that different consensus mechanisms
imply different trust, governance and security assumptions, among other
things, and so cannot be compared to each other simply on the basis of
transactions per second or other performance metrics.
Layer 2 solutions: Application Layer

Another way to improve performance is to move significant parts of
the computation - the work of directing and verifying transactions,
for instance - off the chain into more conventional, and hence more
performant, systems. These are generally referred to as Layer 2 solutions.
A typical Layer 2 solution is a payment channel. In such a case, two parties

who want to transact with each other open a direct channel between
themselves, generally pre-funded, and then carry out their transactions.
At some defined point these transactions are then written to the
blockchain, which acts as the permanent record of what had transpired
— in effect settling the transactions. Such payment channels can greatly
24
increase the performance of a blockchain, but they also imply different

trust assumptions and pose challenges of their own.
More complex variants of this concept can be conceived as well, for

instance multi-sided payment channels or payment hubs, or even
complete sidechains, which are like sub-blockchains that handle
transactions quickly and then write the results to the main chain.1

Up to 10,000 to 100,000 more throughput.
Limitations
This type of solution is the most promising in terms of scalability. It also
often offers interesting tools for privacy. However, it heavily relies on
interoperable tools in order to function properly. When fully mature, layer
2 solutions will allow to connect multiple blockchains seemlessly.
ORACLES
An oracle is an agent that allows to transfer external data to the
blockchain for on-chain use. This is done through the use of smart
contracts that add information about real-world events to the blockchain.
Simple examples of data that are useful to import include temperatures,
prices or information about flight delays. Once entered on the blockchain,
this data can be used to automate processes based on real-world
events (e.g. if a train is delayed, an insurance contract automatically and
autonomously delivers the indemnification).
How does it work?

Technically speaking, oracles are no different from other smart contracts.
However, in order to be useful, oracles need to be trusted: either because
they are operated by a trusted third party or thanks to cryptographic
attestations.
Why is it interesting?
Oracles provide a data feed about external events. Without oracles, a
blockchain would operate only in its own closed environment, limiting
greatly its capabilities.
Maturity and real-world implementations
1 The Lightning Network on Bitcoin and Plasma on Ethereum are examples of sidechains designed to help with scaling.
25
Oracles are proven and easy-to-implement systems.
Limitations
Applications are only as reliable and trusted as their oracles are.
ATOMIC SWAPS
An atomic swap is the exchange of digital assets happening between
assets from two different blockchains, without intermediaries. To take a
simple example, Alice can choose to exchange her digital assets (located
on blockchain A) against Bob’s (located on blockchain B), without relying
on a third party to complete the transaction.
How does it work?

Atomic swaps rely on Hash Timelock Contracts (HTLC) and follow three
main steps:
1. Alice generates a secret key s, calculates its hash h and sends h to
Bob.
2. Alice locks her assets into a contract that Bob can redeem the funds
with s provided within time y or the funds will be sent back to Alice.
Bob then locks his assets into a contract that Alice can redeem the
funds with s provided within time x, such that y is larger than x, or the
funds will be sent back to Bob.
3. Alice reveals s within time x to redeem coins from Bob. Bob then
learns s and can redeem coins from Alice.
Why is it interesting?
Atomic swaps increase the level of interoperability between chains
that are not otherwise linked. They are not specific to a certain type of
blockchain and can theoretically be implemented on any chain that
deals with digital assets.
Maturity and real-world implementations

Atomic swaps is a technology that is already available and was first
experimented with in 2017. It can be used to set up decentralised
exchanges where no third party is required to complete the transaction.
Example: https://komodoplatform.com/.
Limitations
Atomic swaps have so far only been successfully used in order to
exchange simple assets; the technology has not proven its usability to
transfer information.
26
Appendix — Blockchain
Terminology
What is a blockchain?
Blockchain is one of the major technological breakthroughs of
the past decade. A technology that allows large groups of people
and organisations to reach agreement on and permanently record
information without a central authority, it has been recognised as an
important tool for building a fair, inclusive, secure and democratic digital
economy. This has significant implications for how we think about many
of our economic, social and political institutions.
How does it work?

At its core, blockchain is a shared, peer-to-peer database. While there are
currently several different kinds of blockchains in existence, they share
certain functional characteristics. They generally include a means for
nodes on the network to communicate directly with each other. They
have a mechanism for nodes on the network to propose the addition of
information to the database, usually in the form of some transaction, and
a consensus mechanism by which the network can validate what is the
agreed-upon version of the database.
Blockchain gets its name from the fact that data is stored in groups
known as blocks, and that each validated block is cryptographically
sealed to the previous block, forming an ever-growing chain of data.
Instead of being stored in a central location, all the nodes in the network
share an identical copy of the blockchain, continuously updating it as
new valid blocks are added.
What is it used for?

Blockchain is a technology that can be used to decentralise and
automate processes in a large number of contexts. The attributes of
blockchain allow for large numbers of individuals or entities, whether
collaborators or competitors, to come to consensus on information and
immutably store it. For this reason, blockchain has been described as a
‘trust machine’.
27
APPENDIX — BLOCKCHAIN TERMINOLOGY
The potential use cases for blockchain are vast. People are looking
at blockchain technology to disrupt most industries, including from
automotive, banking, education, energy and e-government to healthcare,
insurance, law, music, art, real estate and travel. While blockchain is
definitely not the solution for every problem, smart contract automation
and disintermediation enable reduced costs, lower risks of errors and
fraud and drastically improved speed and experience in many processes.
Glossary
The vocabulary used in the context of blockchains is quite specific and
can be hard to understand. Here are the essential concepts you should
know in order to navigate this breakthrough technology:
• Node: A node is a computer running specific software which allows

that computer to process and communicate pieces of information
to other nodes. In blockchains, each node stores a copy of the
ledger and information is relayed from peer node to peer node until
transmitted to all nodes in the network.
• Signature: Signing a message or a transaction consists in encrypting
data using a pair of asymmetric keys. Asymmetric cryptography
allows someone to interchangeably use one key for encrypting and
the other key for decrypting. Data is encrypted using the private key
and can be decrypted by third-party actors using the public key to
verify the message was sent by the holder of the private key.
• Transaction: Transaction are the most granular piece of information
that can be shared among a blockchain network. They are generated
by users and include information such as the value of the transfer,
address of the receiver and data payload. Before sending a transaction
to the network, a user signs its contents by using a cryptographic
private key. By controlling the validity of signatures, nodes can figure
out who is the sender of a transaction and ensure that the transaction
content has not been manipulated while being transmitted over the
network.
• Hash: A hash is the result of a function that transforms data into a
unique, fixed-length digest that cannot be reversed to produce the
input. It can be viewed as the digital version of a fingerprint, for any
type of data.
• Block: A block is the data structure used in blockchains to group
transactions. In addition to transactions, blocks include other
elements such as the hash of the previous block and a timestamp.
• Smart contract: Smart contracts are pieces of code stored on the
blockchain that will self-execute once deployed, thus leveraging
the trust and security of the blockchain network. They allow users
28
APPENDIX — BLOCKCHAIN TERMINOLOGY
to automate business logic and therefore enhance or completely

redesign business processes and services.
• Token: Tokens are a type of digital asset that can be tracked or
transferred on a blockchain. Tokens are often used as a digital
representation of assets like commodities, stocks and even physical
products. Tokens are also used to incentivise actors in maintaining
and securing blockchain networks.
• Consensus algorithm: Consensus algorithms ensure convergence
towards a single, immutable version of the ledger. They allow actors
on the network to agree on the content recorded on the blockchain,
taking into consideration the fact that some actors can be faulty or
malicious. This can be achieved by various means depending on the
specific needs. The most famous consensus algorithms include proof-
of-work, proof-of-stake and proof-of-authority.
• Validator nodes: Validator nodes are specific nodes in a network that
are responsible for constituting blocks and broadcasting these blocks
with the network. To create a valid new block they have to follow the
exact rules specified by the consensus algorithm.
Learn more about blockchain by watching a recording of our Ask me

Anything session.
29

Report Scalaibility Blockchain

Uploaded by

Copyright:

Report Scalaibility Blockchain

Uploaded by

Document Information

Original Description:

Copyright

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Report Scalaibility Blockchain

Uploaded by

Copyright:

BLOCKCHAIN

About this report

• Members of the Observatory & Forum’s Working Groups.

Scalability Workshop Panelists:

7 Introduction: The blockchain landscape in Europe

A global “backbone” of decentralised chains

10 Main issues and challenges in blockchain

14 Success factors for large-scale blockchain projects

Clear purpose and concept 14

17 Considerations for a European blockchain

Our vision is that the first wave of blockchain adoption will be

• Scalability: the ability to handle large volumes of transactions at high

A lot of time and effort is being expended on surmounting the above

As an increasing number of large projects reach maturity, we can also

We think government can play a role in fostering success as well through

We end with a set of recommendations as well. First priority remains

Introduction: The blockchain

We think this is about to change.

If 2016 was the “year of education”, when people learned about

A by-product of this evolution is a bit more clarity: we begin to see the

INTRODUCTION: THE BLOCKCHAIN LANDSCAPE IN EUROPE

industry or other types of consortia, and focused on a single sector and/or

• Multi-stakeholder-managed industry consortium. A popular

There are many advantages to such market- or use-case-focused

Limiting the blockchain to a single or small set of related use cases

2 For a short overview of blockchain, see Appendix 2 below.

INTRODUCTION: THE BLOCKCHAIN LANDSCAPE IN EUROPE

A GLOBAL “BACKBONE” OF DECENTRALISED CHAINS

Networks are powerful in proportion to their size. As we know, large

Something similar is likely to happen in blockchain.

It seems clear to us that a multiverse of independent blockchains that

This is, however, by no means the only possible scenario. We may

Main issues and challenges in

SCALABILITY That said, compared to conventional financial

MAIN ISSUES AND CHALLENGES IN BLOCKCHAIN

One category of interoperability approaches

MAIN ISSUES AND CHALLENGES IN BLOCKCHAIN

SUSTAINABILITY While these alternatives are all very promising,

1. Environmental sustainability of Those employing proof-of-work also have

MAIN ISSUES AND CHALLENGES IN BLOCKCHAIN

Bitcoin, for example, is a purely open-source

Many of the other significant blockchain

Such projects are often dependent upon

Sustainability can be measured along a

In the next section we analyse some of the

Success factors for large-scale

SUCCESS FACTORS FOR LARGE-SCALE BLOCKCHAIN PROJECTS

SUCCESS FACTORS FOR LARGE-SCALE BLOCKCHAIN PROJECTS

For this reason, successful projects tend to

Considerations for a European

In this section, we offer some general

CONSIDERATIONS FOR A EUROPEAN BLOCKCHAIN INFRASTRUCTURE

CONSIDERATIONS FOR A EUROPEAN BLOCKCHAIN INFRASTRUCTURE

bodies; some of these applications and

to foster innovation. Policy makers should

1 See Blockchain for Government and Public Services, EU Blockchain Observa-

Appendix — Deep Dives

Layer -1 solutions: Hardware Layer

How much scalability can you gain through this layer?

Layer 0 solutions: Network Layer

APPENDIX — DEEP DIVES

propagation such as what data is being sent, how much there is of it

How much scalability can you gain through this layer?