“Proof of work” vs. “Proof of stake” vs. other Byzantine Fault Tolerances
A crucial part of blockchain technology is its ability to produce honesty in an anonymous environment. Since the blockchain is based on consensus, as well as decentralization, truthfulness becomes critical for transactions to occur. Individuals participating on the network need to trust that the balances they hold are accurate. And as blockchain technology has grown in reach, what have also grown are different types of consensus algorithms that provide this source of truth.
In this article, I hope to provide a summary on the various consensus algorithms used by blockchains. These varying algorithms have two primary aims;
1. Creating a single version of truth, and
2. Preventing bad actors from committing fraud
To create a secure consensus environment, it must be fault tolerant, since there is no central authority to reconcile differences. The following algorithms are used to achieve Byzantine Fault Tolerances that mitigate false transactions and promote honesty in a decentralized space.
“Proof of Work” method
The “proof of work” method is the consensus algorithm used by Bitcoin and other popular cryptocurrencies. It is the most highly used consensus algorithm. In simple terms it involves having computers solve a challenging puzzle that solidifies all transactions in the next new block. The puzzle is used to slow a would-be attacker down and make the verification process more likely to be conducted by honest members of the network. Honest members are rewarded for accurate verification while the network can certify the new block before the next block is created.
The puzzle that these honest participants are solving requires near endless guess and check. That’s one of the reasons why these actors are called ‘miners’ because they are constantly hammering away with guesses into the system to see if the guesses are sufficient. As depicted in a prior post, miners are aggregating information associated with this new block and are linking this to the blocks before. The last piece of information to form this link and move onto the next block’s new transactions is the ‘nonce’, or the answer that miners are looking to solve.
What I’ve read, is that the ‘nonce’ is a bunch of gibberish a computer is looking for to transform the block’s hash to start with a certain number of zeros (Somewhere around 18 zeros in Bitcoin’s blockchain)[1]. The nonce is a portion of the input that is used to solve the problem miners are working towards. It can take some of the fastest computers in the world, that are completely dedicated to mining operations, around 10 minutes to accomplish this puzzle. For bad actors, the difficulty level makes it very challenging to submit a bad block with fraudulent transactions. And since blocks are built from one another, for a bad actor to try and change prior blocks it would have to change the nonce of one block and do that for all the subsequent blocks thereafter (Which would be nearly impossible).
Once the miner completes the puzzle, the miner broadcasts the accomplishment to the network. Other miners verify the answer and make sure that the block includes valid transactions. Then miners begin working on a new puzzle based on the results of the completed block. This is the “proof of work” concept that forms the foundation of consensus for Bitcoin and other alternative coins.
One major issue with this method of consensus is the amount of computing power required to perform the mining function. From Newsweek, Bitcoin mining consumes 33 terawatts of electricity annually (Greater than 159 individual countries)[2]. As more computers go online and focus on mining, the “proof of work” puzzle will become equivalently more challenging (Keeping new block generation around 10 minutes). Fewer and fewer groups will be able to support such mining operations. The necessary function for mining and providing ledger consensus may end up as the responsibility of a few individual entities that have the resources and computer power to mine these functions. This scenario would then be the opposite of the original goal for a distributed anonymous ledger, set forth by Bitcoin.
A cool website to see the ever-increasing energy consumption of Bitcoin.
“Proof of Stake” method
“Proof of stake” method is another consensus algorithm, currently being utilized by ShadowCash, Nxt, BlackCoin and Qora. It is also being strongly considered by Ethereum. “Proof of stake” differs from “proof of work”, in that “proof of work” has thousands of computers competing to solve a complex puzzle for a monetary incentive. For “proof of stake”, the creator of the new block is chosen based on the percentage of the overall coin supply that the creator owns[3]. In other words, if an individual on the network owned 10% of the coins in the blockchain then that person has a 10% chance of being the next block creator.
If you are the creator and you assemble all the transactions and compile the block’s pieces truthfully, you are not given any new coins. All the currency has already been created at the beginning of the blockchain (The number of coins won’t change). However, if you are the creator and you do a good job you are given transaction fees. Once the block is completed, it is audited by member of the network for verification and added to the chain.
If you happen to be naughty and include fake transactions or alter the transactions in the block you built, then you’ll risk losing all the coins in your possession. You earn a small incentive from the transaction fees if you are truthful, but lose everything if you are dishonest
“Proof of stake” greatly reduces the energy consumption issue associated with computing power. Only a few computers are working and verifying the transactions for each block, which leads to enormous energy savings compared to “Proof of work.”
“Proof of stake” however, though efficient, doesn’t have the same amount of attention and legacy timespan that “Proof of work” has. It has not been able to show that it can defend itself against malicious attacks and bad actors.
“Proof of work” and “Proof of stake” are by far the most popular consensus algorithms used by blockchains. The following algorithms I’ve stumbled upon in my research are less used but are all still related to solving Byzantine Fault Tolerances. I have provided a brief description with links for further content.
This method is a combination of both “Proof of work” and “Proof of stake.” From my understanding, participants start out by mining blocks. The winning block is then signed by validators, who are chosen by the number of coins that they own (Owning more coins, you are more likely to be selected).
This method has participants burn coins to have the privilege to mine based on a random selection process. Overtime, you would need to burn more and more coins to have the privilege to mine.
This method doesn’t burn nearly as much energy as “Proof of work” method. However, like “Proof of work” method, “Proof of burn” is using resources inefficiently and will most likely lead to a few individual contributors (Those wealthy enough to continue to purchase more coins to have the mining privilege) who oversee transcribing the ledger.
This method is like “Proof of work,” but instead of computation usage, computer storage is used. “The mining algorithm is too complicated to calculate in real time, and the block times are so short (an average of 1 block every 4 minutes), that solutions must be saved on the hard drive ahead of time. Ultimately, the more solutions, also called plots, you have on your hard drive, the greater your chance of having the best solution for the most recent puzzle.[4]”
This method prevents the usage of energy for transcribing on the ledger. However, it still leads to individuals who devote more resources to large amounts of storage, becoming the minority who control transcribing the ledger. Not an outcome desired for a consensus algorithm.
For more of my writing, please check out: http://ibsix.com/
[1] Norman, Alan T. Blockchain Technology Explained: the Ultimate Beginner’s Guide about Blockchain Wallet, Mining, Bitcoin, Ethereum, Litecoin, Zcash, Monero, Ripple, Dash, Iota and Smart Contracts. Alan T. Norman, 2017.
[2] Cuthbertson, Anthony. “Bitcoin’s Meteoric Rise Is Very Bad News for the Environment.” Newsweek, 11 Dec. 2017, www.newsweek.com/bitcoin-mining-track-consume-worlds-energy-2020-744036.
[3] Norman, Alan T. Blockchain Technology Explained: the Ultimate Beginner’s Guide about Blockchain Wallet, Mining, Bitcoin, Ethereum, Litecoin, Zcash, Monero, Ripple, Dash, Iota and Smart Contracts. Alan T. Norman, 2017.
[4] “What Is Proof of Capacity? An Eco-Friendly Mining Solution.” CoinCentral, 1 Feb. 2018, coincentral.com/what-is-proof-of-capacity/.