bitpro가 원문에서 글자만 모은 내용이에요. 표·그림·수식은 빠지거나 깨질 수 있어요. 전체 5쪽 중 1쪽
Conflux Network:
Engineering An Economic Design
Andreas Park1,2 and Andreas Veneris1,3,4
1Conflux Foundation Advisor
2The Rotman School of Management, University of Toronto
3Dept. of Electrical and Computer Engineering, University of Toronto
4Dept. of Computer Science, University of Toronto
andreas.park@rotman.utoronto.ca, veneris@eecg.toronto.edu
Abstract
Distributed ledgers or blockchains, in particular those based on Proof-of-Work (PoW)
protocols, rely as much on economic mechanisms as on technology. In this paper, we
describe how we approach the design of the economic mechanisms that underline Con-
flux, a high throughput/performance PoW blockchain, by providing a detailed analysis
of its economic viability. Conflux offers several innovations relative to well-understood
blockchain networks such as Bitcoin and Ethereum, both in terms of technology design
but also in terms of the economics that underpin the technology. Most notably, a ma-
jor difference in Conflux when compared to the status-quo is that processing of blocks
occurs in parallel rather than serially, and users who commit code or information to the
blockchain face ongoing costs and/or benefits for the duration they occupy chain space.
Hence, the ecosystem design requires to assess carefully those parameters including how
new block processing affects the fundamental limits of blockchain performance, and how
storage costs affect user incentives in this new decentralized ecosystem.
1
NOTICE
NOTHING IN THIS WHITEPAPER CONSTITUTES LEGAL, FINANCIAL, BUSI-
NESS, OR TAX ADVICE AND YOU SHOULD CONSULT YOUR OWN LEGAL,
FINANCIAL, TAX OR OTHER PROFESSIONAL ADVISER BEFORE ENGAGING
IN ANY ACTIVITY IN CONNECTION HEREWITH. NEITHER CONFLUX FOUN-
DATION LTD. (THE CONFLUX), ANY OF THE PROJECT TEAM MEMBERS
WHO HAVE WORKED ON THE CONFLUX PLATFORM OR PROJECT IN ANY
WAY WHATSOEVER (THE CONFLUX TEAM) NOR ANY THIRD PARTY SER-
VICE PROVIDER SHALL BE LIABLE FOR ANY KIND OF DIRECT OR INDI-
RECT DAMAGE OR LOSS WHATSOEVER WHICH YOU MAY SUFFER IN CON-
NECTION WITH ACCESSING THIS WHITEPAPER, MATERIALS PRODUCED BY
THECONFLUX,ORACCESSINGTHEWEBSITEAT[HTTPS://WWW.CONFLUX-
CHAIN.ORG/] OR ANY OTHER MATERIALS PUBLISHED BY THE CONFLUX.
The Conflux and the Conflux team do not and do not purport to make, and hereby
disclaims, all representations, warranties or undertaking to any entity or person. All
statements contained in this Whitepaper, statements made in press releases or in any
place accessible by the public and oral statements that may be made by the Conflux
and/or the Conflux team may constitute forward looking statements (including state-
ments regarding intent, belief or current expectations with respect to market conditions,
business strategy and plans, financial condition, specific provisions and risk manage-
ment practices). You are cautioned not to place undue reliance on these forward-looking
statements given that these statements involve known and unknown risks, uncertainties
and other factors that may cause the actual future results to be materially different from
that described by such forward looking statements. These forward-looking statements
are applicable only as of the date of this Whitepaper and the Conflux and the Con-
flux team expressly disclaims any responsibility (whether express or implied) to release
any revisions to these forward-looking statements to reflect events after such date. You
understand that the Project and the creation and distribution of the Tokens involve
significant risks, including but not limited to, the risk that (i) the technology associated
with the Conflux Project may not function as intended; (ii) the Conflux Project may fail
to attract interest or adoption, either from key stakeholders or the broader community;
(iii) no guarantees that the price per Token determined by the market will be equal to
or higher.
1
I. Introduction
When NASA prepared to travel to space they spent millions to develop a ballpen
so that astronauts can write in zero gravity. Faced with the same problem, the Soviet
Union used pencils. This anecdote is Conflux’s philosophy in a nutshell: its main goal
is to take the best parts of existing permissionless blockchains and push those bounds
significantly.
What is Conflux? In brief, it is a Proof-of-Work (PoW) blockchain network that
allows the parallel processing of blocks and transactions, while eventually forming a
final serial chain. This feature contrasts other well-known networks, such as Bitcoin and
Ethereum, which process blocks strictly one-by-one.1 The parallel processing creates
economic incentives for miners that are notably different from serial chains and in this
paper we discuss how the differences improve the security and economic viability against
double-spending attacks.
Conflux also seeks to overcome the problem of the unpaid smart-contract space occu-
pation. In Ethereum, when introducing a new contract, a user merely pays at the time
of the inclusion of the code in the chain whereas the community as a whole faces the
burden of keeping the contract in storage. Conflux introduces ongoing costs of contract
maintenance which give users an economic incentive to avoid wasting resources.
Toprovideintrinsicvalue, Confluxseekstoattractusersthatactivelyusethenetwork
for value-added services. Conflux believes it is critical that there are no barriers-to-entry
for those who make meaningful contributions. This is why Conflux is a permission-
less network with economic-driven governance mechanisms that incentivize community
1Ethereum does allow a limited number of so-called “uncle” blocks to allow for faster processing, but
the setup of Ethereum is intrinsically not parallel.
2
contributions to ensure a secure, stable, and predictable environment for commercial
activities.
ThisdocumentdiscussesseveralaspectsoftheConfluxnetworkthatinvolveeconomic
mechanisms. These mechanisms affect the users’ and miners’ costs and rewards and thus
their behavior, therefore it is crucial to plan ahead so as to avoid creating unintended
consequence. This paper is a companion paper to the technical Conflux manuscript. It
complements the engineering aspects by focusing on the underlying system’s micro- and
macro-economic mechanisms.
At its core, blockchain technology is an infrastructure solution that allows the secure
(i) transfer of economic value and(ii) execution of programming state/storage, without
a trusted third party. To succeed in this scope, it needs to balance several core factors.
First, economic value transfers and payments for the cost of such usage rely on a
native token. This token needs to be designed so as to serve as a medium of exchange
but also as a unit of account for blockchain-based data. It is also desirable that the token
canbeusedasastoreofvaluesoitcansustainlong-runincentivesforthedifferentparties
to use the chain. Finally, as is common with PoW networks, the native token plays a
role in the compensation of the network’s miners who are central to the operation and
security of the network. There are by now a number of papers that highlight how native
tokens can spur the adoption and usage of a network,2 and we build on this body of
literature in our analysis.
In this document we describe the distribution of tokens as well as the usage rules
and the economic impact of the various rules. Namely, tokens are issued at genesis,
as interest payments on subsets of tokens that satisfy certain rules, and as rewards to
2See, for instance, Bakos and Halaburda (2018), Cong, Li, and Wang (2018), Fisch (2019), Canidio
(2018), or Li and Mann (2018)).
3
miners who include new blocks. We then provide a calibrated model that simulates
the miners’ expected income. Tokens interact with the outside world in the sense that
there is a exchange rate to fiat currencies (as well as to other crypto-currencies), and we
therefore also discuss carefully how the issuance of tokens affects the exchange rates.
In the final part of the paper we provide two formal analyses to provide a better un-
derstanding of the economics of Conflux. First, there are some well-known constraints
on double-spending proofness for PoW blockchains3 and we highlight how the Conflux
Network expands the set of economically viable states relative to known networks. Sec-
ond, we provide a formal equilibrium model of user and miner interaction to study how
changes in policy variables affect equilibrium outcomes.
II. Cornerstones of the Existing Blockchain World
Before presenting the proposed economic system, we review key concepts from the
main existing blockchain models.
Bitcoin is the first working cryptocurrency that solved the double-spending problem.
At its core, the Bitcoin blockchain is a one-trick pony that is designed for the transfer
of its native token, Bitcoins. It is conceptually set up to be slow as blocks of fixed size
are produced serially at a approximately constant rate of one block, which can contain
only around 3, 500 transactions, per 10 minutes. Although for a good level of certainty
that the value transfer has indeed happened, it is not sufficient that a transaction has
been included in a block. The reason is that the block may not be a part of the longest
chain but of a fork. This can happen, for instance, when two blocks are created at
around the same time — only one of them will eventually become part of the longest
3See Chiu and Koeppl (2017) and Budish (2018).
4
chain. One therefore needs to wait for a few block confirmations before the transaction
can be considered as final. The setup, with block formation time of 10 minutes and a
even longer confirmation time, is clearly unsuitable for day to day payments. Bitcoin
has also other limitations: once all Bitcoins have been mined through block rewards,
miners receive only user-determined transactions fees for securing the network. If we
take the current income of miners as a benchmark, these fees need to be rather large:
At the current block rewards of 6.25 BTC per block and a BTC price of approximately
$10,000 (as of the day of this publication)each of the 3,500 transactions that fit into
a block needs pay a fee of around $18. All these features make Bitcoin unsuitable for
remittance payments and likely too expensive to be a substitute payment network for
the majority of business transactions (Auer 2019).
The main participants in the Bitcoin network, however, are the miners, short-term
speculators, and long-term holders. Very few network participants use Bitcoin for its
original purpose, as “peer-to-peer electronic cash.” In the future, the long-term holders
need to rely on short-term speculators to create sufficient transactions so that the miners
are willing to continue to provide the security for the network. This will likely become a
challenge in the future. In other words, as it stands, the Bitcoin network today does not
incentivize users of the token for meaningful and value-generating economic activity. In
addition, the technology infrastructure of Bitcoin has limitations and allows little beyond
transfers of Bitcoin. Without Turing-complete scripts, decentralized application (dApp)
developers have at most peripheral use for it through introduction of “side-chains” such
as the Lightning Network, etc. Governance is also a challenge. Changes to the Bitcoin
protocol require that miners agree. This can be difficult when those changes may have
5
a negative impact on the miners’ income — even if the change itself serves the broader
ecosystem well.
The Ethereum Network is a much improved system relative to Bitcoin and features
Turing-complete smart contract functionality, which enables developers to code dApps
that run on the network. Ethereum’s smart contract language allows the creation of
tokens that can be used as non-native payment coins or even as securities. Ethereum’s
native token Ether has no hard cap on its token supply and can therefore grow organ-
ically as usage of the network expands and demand for tokens increases. Ethereum’s
governance follows a similar mechanism as Bitcoin, with some improvements: miners
receive not only block rewards for each block that they create, they also collect transac-
tion fees for the execution of code. Therefore, more complex transactions or operations
are rewarded with higher fees.
However, as in Bitcoin, Ethereum also suffers from a low transaction throughput
rate: the network can currently process only a maximum around 40 transactions per
second. This is nowhere near the rate of need for modern commercial financial infras-
tructure. For instance, Payments Canada processes around 28 million transactions on
an average business day, while Canadian economy is ranked only the10th by Nomi-
nal GDP (Times 2019). Assuming that most activities occur over 12 hours daily, a
blockchain payments network to handle this level of transactions would need an approx-
imately minimal throughput of 650 transactions per second (tps). Furthermore, it is
equally infeasible to use the Ethereum blockchain as a substitute infrastructure for secu-
rities trading: traditional centralized securities exchanges and over-the-counter securities
markets worldwide process billions of orders and transactions per day. As a matter of
fact, although there are many decentralized exchanges on the Ethereum network, at the
6
current throughput rates they are fundamentally unable to accommodate a meaningful
fraction of today’s financial market activities.
A second function issue for Ethereum is that users pay for a contract only at the
time of the inclusion of the contract code. Additionally, only the miner that includes
this code on the chain gets this one-time reward. Yet every new contract submitted to
the Ethereum chain not only requires the network to execute the code, the contract also
occupies “chain-space” (i.e., memory in the global state of the Ethereum chain) even
if it remains inactive after its submission. In summary, all full nodes of the network
need to subsequently store the information resulting in a situation where users can store
data on the chain with a one-time inclusion fee, while the storing of such data can
be indefinitely long with no maintenance payment required. As such today, the vast
majority of the state-tree space of Ethereum is occupied by inactive smart contracts.
This unused data waste space in the state-tree of the blockchain, slow down the system,
and create undesirable network latency/overhead.
One of the goals for the Conflux project is to build on and vastly improve on the
features of existing blockchain systems: guaranteed execution of software code via a
Turing-complete virtual machine on a permissionless consensus mechanism with a higher
performance while not sacrificing safety and decentralization.
III. An Overview of the Conflux Network
Conflux is a new PoW network with a Turing-complete smart contract language
similar to this of Ethereum.4 The Conflux network provides significant performance
improvements with its processing of parallel blocks in a Directed Acyclic Graph (DAG)
4See Li, Li, Zhou, Xu, Long, and Yao (2018) and .
7
structure, which lowers confirmation times and increases transaction throughput sub-
stantially.
To address the space congestion challenge, Conflux requires users to bond native to-
kens into storage to occupy space, which implicitly creates a disincentive to occupy space
unnecessarily. The disincentive stems from the payment of interest on existing tokens in
the system. The interest on the bonded storage is payed to miners instead of the users to
create a long term income to the miners. To address the fairness attack challenge, Con-
flux assigns the block reward in a way that eliminates the winner-take-all characteristic
of mining. Instead of competing for the longest chain, miners in Conflux receive block
rewards for all the blocks that they generate, albeit with some penalty mechanisms that
encourage following the consensus protocol. Competing blocks are jointly penalized so
that selfish mining is not profitable and different miners are incentivized to cooperate
along the protocol to keep the network stable and secure.
Similarly to Ethereum, Conflux operates with an account-based model that every
normal account associates with a balance and each smart contract account contains
the corresponding byte codes as well as an internal state. Conflux supports a modified
version of Solidity (the main contract language in Ethereum) and Ethereum Virtual
Machine (EVM) for its smart contracts, so that smart contracts from Ethereum can
migrate to Conflux easily.
A transaction in Conflux refers to a message that initiates a payment transaction, or
deploys/executes smart contract code. Each block consists of a list of transactions that
are verified by the proposing miner. Each node maintains a pool of verified, received
transactions that have not yet been included in a block. Miners compete with one an-
other by solving PoW puzzles to include transactions into blocks. Similar to Bitcoin and
8
Ethereum, Conflux adjusts the PoW difficulty so as to maintain a stable block generation
rate. Each node also maintains a local state constructed from the received blocks.
The Conflux consensus algorithm operates with a special Directed Acyclic Graph
(DAG) structure called TreeGraph. Unlike Ethereum which only accepts transactions
on a single chain into its ledger, the Conflux consensus algorithm safely incorporates and
processes transactions in all concurrent blocks. There are two kinds of edges between
blocks, parent edges and referenceedges. Each block (except the genesis) in the Tree-
Graph has exactly one parent edge to its chosen parent block. Each block can also have
multiple reference edges to refer previous blocks. All parent edges form a tree embedded
inside a Directed Acyclic Graph (DAG) of all edges.
At a high level, Conflux uses the novel Greedy Heaviest Adaptive SubTree (GHAST)
algorithm (Li and Yang (2020)), which assigns a weight to each block according to the
topologies in the TreeGraph. Under this weight assignment, there is a deterministically
heaviest chain within the graph calledpivot chain, which corresponds to the relatively
most stable chain from the genesis to the tip of the parental tree.
Parent edges, reference edges, and the pivot chain together enable Conflux to split
all blocks in a DAG intoepochs. As shown in Figure 1, every block in the pivot chain
corresponds to one epoch. Each epoch contains all blocks that are reachable from the
corresponding block in the pivot chain via the combination of parent edges and refer-
ence edges and that are not included in previous epochs. Details about the consensus
algorithm can be find in (Li and Yang (2020)).
Experimental results have shown that Conflux is capable of processing4, 000 trans-
actions per second for simple payment transactions, at least two orders of magnitude
higher throughput than Ethereum and Bitcoin. The improvement in throughput is a
9
result of the TreeGraph structure and the consensus algorithm, so that the network can
operate with a much faster block generation rate, no forks are discarded, and with a
higher utilization of block space. According to the technical specification, the main net
of Conflux will run under a fixed block generation rate at two blocks per second. The
daily block generation rate is therefore60· 60· 24× 2 = 172, 800 blocks per day.
IV. Token Rules
There is a unique native token on the Conflux network, hereafter referred to asCFX.
Each CFX contains1018 Drip. Transactions on Conflux are handled similarly to those
on the Ethereum network and CFX, therefore, plays a similar role as Ether. Namely,
users submit transcation with a gas limit and a gas price; the latter is denominated
in CFX.
The Conflux Foundation is a non-profit entity established by Conflux to make ad-
justments when the allocation of resources deviates from an equilibrium, to provide in-
centives to overcome the cold start problem, and to promote participation/development
in the network in the early stage.
A. Genesis Tokens
The initial number of tokens is 5,000,000,00프로젝트로 돌아가기