Show me what you got, Proof-of-Stake
DSRV Research publishes in-depth blockchain-related content with the aim of safely onboarding more people to Web3.
Disclaimer: This article is for informational purposes only and should not be taken as financial advice. No information contained within this article is a recommendation to invest in any of the assets mentioned. All investors are advised to thoroughly conduct their own research before making any financial decisions.
Proof-of-Stake: The Quest For a Better Consensus Algorithm
[Consensus Algorithm Series]
- Boom Boom PoW, Proof-of-Work
- Show Me What You Got, Proof-of-Stake
As Bitcoin and Ethereum edge their way into the mainstream, who stands to benefit the most? The person who randomly bought some bitcoin and became unknowingly rich? The one who periodically mines Ethereum while running an internet cafe? Or the lucky mining equipment manufacturer whose business took off with the rise of BTC and ETH (not to mention the consistent demand as mining rigs are vulnerable to humidity and outdated ventilation systems)?
Because blockchain is essentially a distributed network run by multiple computers, there must be a way to agree upon the validity of data before it is stored by all network participants. This mechanism is known as a consensus algorithm. The two most well-known cryptocurrencies, Bitcoin and Ethereum, both use Proof-of-Work (PoW) as their consensus algorithms. In order to mine crypto on a PoW blockchain, miners must race to solve a sophisticated mathematical problem. And while this contributes to a very secure infrastructure, as the mining competition intensifies, there is an incredible surge in both energy and electricity consumption.
While PoW-based cryptocurrencies are not the only factor driving GPU demand, the correlation is still definitely worth noting. Miners are incentivized to acquire better and more powerful GPUs because it increases their ability to mine successfully. And this in turn raises a fundamental question: Why do we need this much electricity and equipment just to decide who’s going to mine the next block? What began, for some, as a side hustle to make some extra money is quickly becoming an insanely expensive endeavor.
Enter Proof-of-Stake (PoS). Swap out the expensive mining equipment for a sizable number of coins, and you have a shot at mining the next block. No more maintaining heavy-duty equipment, no more exorbitant electricity bills. If more people can easily create and distribute cryptocurrency in this way, do we get closer to achieving blockchain’s ethos of decentralization?
Let’s take a closer look!
The Disadvantages of Proof-of-Work
Excessive Energy Consumption
While Proof-of-Work has long proven itself to be incredibly secure, it now entails considerable financial and environmental costs. Excessive energy consumption is behind most of the criticism leveled against Proof-of-Work, which is unsurprising given that Bitcoin consumes more electricity per year than Ukraine or Norway [1]. This includes the recent EU regulations on cryptocurrencies that zero in on the environmental sustainability of PoW chains, specifically the concerns surrounding greenhouse gas emissions caused by excessive energy consumption [2]. Some argue that Bitcoin doesn’t significantly affect the environment as over 60% of the Bitcoin mining industry uses renewable energy [3]. And yet, PoW chains still consume far more energy than PoS ones do due to the competitive, problem-solving nature of the consensus algorithm. This mechanism is unique to PoW blockchains.
Excessive E-Waste
Excessive e-waste (electronic waste) is another common criticism of Proof-of-Work. In the early days, miners could get away with only using CPUs, but as the competition intensified, GPUs and even mining-specific hardware known as ASICs (https://www.investopedia.com/terms/a/asic.asp) became the norm. As hardware is continuously innovated over time, outdated equipment loses its value and falls behind in the mining market. And it doesn’t last forever either, so e-waste generation is ultimately inevitable.
Centralization Concerns
Centralization is also an issue. Ironically, the centralization in PoW networks is caused by its high degree of decentralization; as technically anyone can participate in the network as a miner, competition intensified and rendered CPU or GPU mining almost obsolete. ASIC chips newly developed for the purpose of mining Bitcoin entered the market, effectively commercializing Bitcoin mining. Currently, Bitcoin has several different mining pools as shown below.
A mining pool is simply a group of miners that combine their resources in order to increase their chances of mining a block. This has raised the bar for individual miners, who are now faced with substantially higher barriers to entry. It’s worth considering whether this is at all in line with what Satoshi Nakamoto wanted when he first introduced Bitcoin as a decentralized electronic currency.
We can also look at this centralization issue geographically. In order to maintain a profitable mining business, it goes without saying that one should prioritize countries with cheap electricity when building mining farms. This means Bitcoin is now more vulnerable to conditions in these specific locations — natural disasters, blackouts, cryptocurrency regulations, and the like — instead of existing as a truly dispersed, global system free from geographical/ political constraints.
A case in point was the June 2021 dip in Bitcoin hash power (total computing power), after miners based in China were forced to close up shop due to the country’s crackdown on mining. A drop in hash power poses a major security threat because it increases the risk of hacking– this is also an unforeseen problem caused by excessive centralization.
Next Up, Proof-of-Stake
By now we’ve set the stage for PoW’s successor, Proof-of-Stake. Unlike Proof-of-Work, Proof-of-Stake does not rely on specialized hardware to maintain network security, thereby enabling a wider array of people earn block rewards. This already improves on both decentralization and sustainability. Ethereum is currently making the transition from PoW to PoS for these very reasons. And as the second largest blockchain by market cap, Ethereum’s move signals a larger industry paradigm shift from PoW to PoS.
Because each implementation of PoS boasts a unique architecture and set of features, it’s fairly hard to give a single definition that fits all. Nevertheless, we can broadly define PoS as a mechanism that selects participants to validate the next block based on how many tokens they stake (pledge to the network.) What exactly does this mean? Let’s have a look at some examples.
Let’s assume that there’s a blockchain with a total of 10,000 tokens in circulation. Validators A, B, and C participate in block production and own, respectively, 100, 300, and 600 tokens. In order to participate in the block production and validation process, each validator must stake their tokens as collateral. If A, B, and C participate by staking all of their tokens, the total number of tokens involved in the block production process is 1000. This means A holds 100 out of 1000 shares, or 1/10 of the total amount staked to the network. As a result, the probability that A produces the next block is also 1/10 (10%).
Thus, in order to increase their chances of producing the next block, validators must simply acquire more stake rather than mining equipment or electricity. The fact that we no longer need mining equipment is especially significant, because this kind of capital equipment helps businesses achieve economies of scale [4]. To get involved in the PoW mining business, you need land, property, and enough electricity to install and operate hardware. Companies that have been in the mining business for several years can immediately improve their odds by acquiring, say, $10k worth of more hardware. But for an individual who wants to start mining now, $10k will not even be sufficient to buy a large enough plot of land. This means Proof-of-Work is relatively vulnerable to centralization, as big players can use additional capital more efficiently than less established individuals.
Of course, PoS isn’t without its flaws either. Because there’s no need to buy mining equipment, individuals with a small stake can still participate in block production. This can result in an extremely high number of nodes joining PoS networks– most of which do not set a maximum number of validators. And if too many nodes join the network, the time and network bandwidth needed to reach consensus and validate a block will increase significantly. We can see this in the image below: while it takes 3 rounds of communication for three validators to reach consensus, it takes 6 rounds for six validators.
Therefore, the greater the number of validators, the more time it takes to produce a block, resulting in a slower processing speed overall.
Delegated Proof-of-Stake (DPoS)
Decentralization, or the lack thereof, is clearly an issue. But what if, instead of wholly sacrificing decentralization in favor of faster processing times, we could manage centralized validators in a relatively decentralized manner? This is essentially what Delegated Proof-of-Stake (DPoS) was created to do. Instead of strictly limiting participation in block production to a set number of validators, DPoS allows users to delegate their stake to validators and earn a share of the block reward.
Another way to think of DPoS is as a representative democracy. If citizens had to cast votes on all matters, not just the presidential and parliamentary elections, they’d be making trips to polling stations all year round. By electing an official to represent the people, the entire system can function much more efficiently. While DPoS limits the number of validators that can participate in block production, there is no limit on the number of users that can delegate to these chosen validators.
Let’s go back to our original example featuring validators A, B, and C to look at how the DPoS structure works in practice. In the case of a basic PoS chain, token holders like D, E, and F could simply participate as validators for the network. On a DPoS chain with a three-validator limit, D, E, and F will instead delegate their stake to A, B, and C. Let’s say that D, E, and F have 50 tokens each. Assuming that the nodes (computers) managed by validators A, B, and C are all operating well, they may decide to delegate their combined 150 tokens to A, who happens to be the validator with the lowest stake. A will now have a total of 250 tokens, which increases A’s probability of producing the next block from 10% (10/ 1000) to 21% (250/ 1150).
Thanks to token delegation, A can earn more block rewards than in regular PoS, while C earns fewer than before even with the same stake. Of course, D, E, and F delegated to A receive a block reward from A for the amount they delegated, excluding the fees. And D, E, and F can in turn easily earn block rewards through delegation rather than sophisticated node management. This is how we approach a centralized architecture in a decentralized fashion.
Contrary to a representative democracy in which officials are elected for a fixed term, DPoS validators are subject to redelegation (unstaking from one validator and delegating to another) if they act dishonestly or fail to properly manage their nodes. These economic incentives therefore ensure that validators behave honestly. In the case of more extreme transgressions, validators’ stake can also be slashed (a.k.a. confiscated by the network), along with that of their delegators– a reminder that choosing a good validator is essential!
Wrap-Up
So far, we’ve covered the reasons behind the transition from Proof-of-Work to Proof-of-Stake, as well as the difference between the latter and Delegated Proof-of-Stake. While each new iteration brings new solutions to the table, there is no single best consensus algorithm.
Instead, we can think of PoS and DPoS as belonging to a continuously evolving spectrum, rather than as isolated concepts. On one end is standard PoS that seeks to improve on decentralization, on the other is a single-validator DPoS with an emphasis on processing speed above all else. When creating a blockchain, the problem you want to solve most will determine where you fall on the spectrum. Say you want to launch a fast payment system using blockchain technology. How much decentralization should you sacrifice for the processing speed you want? Should you limit the number of validators to 100? Or 1000? These are the questions to consider.
Hopefully this framework will help you better understand why existing blockchains made the design choices they did (and assess for yourself whether these choices have achieved their purpose!). There are many efforts currently being made to solve the dilemma of decentralization versus processing speed. Ethereum, specifically, introduced a novel concept called ‘sharding’, which is essentially the process of splitting a database horizontally to spread the load. We’ll get into this in upcoming articles, so stay tuned!
Written by
Seokjoong Yoon, DSRV Research Associate (Twitter @imlearning_eth)
Reviewed by
Owen Yoonjae Hwang, DSRV Research Manager (Twitter @journeywith_eth)
Youngbin Park, DSRV Research Associate (Twitter @bin0_0bin)
Translated by
Domitille Colin, Communications Manager (Twitter @domitille_marie)
Illustrations by
Heeyoung Moon, Brand Designer
💊 Key Takeaways
- Proof-of-Work (PoW) grapples with sustainability issues due to excessive energy consumption, e-waste generation, and centralization.
- Proof-of-Stake (PoS) was conceived as a response to some of PoW’s core problems.
- A variation of PoS, Delegated Proof-of-Stake (DPoS) is characterized by a limited number of validators, making it more centralized but also much faster than regular PoS.
References:
[1] Cambridge Bitcoin Electricity Consumption Index
[2] Environmental sustainability key sticking point in EU MiCA bill — Cointelegraph
[3] Bitcoin Mining Council Survey Confirms Sustainable Power Mix
[4] Proof of Stake FAQ — Vitalik Buterin
