In order to more easily mine bitcoin (BTC), miners most often group themselves into pools. This grouping essentially allows them to pool efforts to have better resolution power. However, the importance of a pool can be devastating for the entire community. One of the most feared consequences in this case is nothing more than a 51% attack. But what is it really about? What are the consequences of such an attack and how is it carried out? The content of this article will surely allow you to see more clearly.
A hostile takeover
The 51% attack consists of a hostile blockchain takeover by a group of individuals holding the majority of the computing power of the network. Attackers can therefore cancel transactions made by others and decide which ones will be validated. This malicious attack targets a Pow type blockchain (Proof of Work). It is actually this mode of governance that makes attack possible and makes it the perfect target. Indeed, in this decentralized governance, the transaction register is updated in depending on what the majority approves.
Result, this is the longest blockchain that is saved and shared, because it is representative of the majority of minors. After all, it takes the majority of the computing power to create the longest version of the chain. This allows in particular that a modified block is automatically rejected by the majority of the network. This measure, however, only prevents isolated attacks and small attempts to hijack the blockchain. If an ill-intentioned mining pool had this power, it could easily carry out the 51% attack.
Presentation of the different stages of a 51% attack on Bitcoin
To perfectly illustrate the course of a 51% attack, we will consider a mining entity X deciding to attack the Bitcoin blockchain. Of course, this hostile entity must hold the majority of the computing power of the network.
It all starts with the resolution of a hash by entity X for the formation of a block. However, this new block formed will not be immediately communicated to the other miners for verification and insertion into the chain. Rather, it will serve as a basis for start another parallel chain which will only be known to entity X. So we have a public channel (the real one) and a illicit chain that secretly forms.
Entity X now goes complete a transaction which will be shared on the public channel. This transaction will consist of a purchase of goods against a certain amount of bitcoins. This transaction will approved and registered on the public blockchain. However, it will not appear in the register of the illegal channel which continues to grow in secret. The validation being made on the public channel, the good will be delivered to the entity X. The falsification of the data being successful, the entity X will now deploy all its computing power so that its fictitious chain is the longest.
Once the channel extension is finished, our new entity will now distribute this version to the rest of the network. If the available computing power is sufficient, this new channel will be considered legitimate, because being the longest and supported by the majority of the network. The previous transaction made was not recorded on it, entity X has both the good and the bitcoins that are supposed to have bought it. It can therefore reuse them for other purposes, hence the notion of Double Spending (double spend) the other name for the 51% attack.
Consequences and feasibility of a 51% attack on a blockchain
The first consequence of such an attack is that the asset that was the subject of this double expenditure will lose value. Its reputation will be compromised, and it will be used less and less to carry out transactions. In a case other than that of cryptocurrency, attacking the 51% on a blockchain will simply allow falsify information content stored there. While the 51% attack is theoretically easy to do as long as the conditions are right, the reality is quite different.
Indeed, it would be really difficult, or even impossible for a minor, to deploy the arsenal necessary to have sufficient computing power giving him the majority of the voices of the network. This is particularly true on Bitcoin’s blockchain, the oldest, and mechanically the most robust, so much so that gathering 51% of the computing power is virtually impossible (and / or incredibly expensive).
The logistical and financial constraints that this implies seem insurmountable. In the event that a mining pool with the necessary resources considers it, it would be the first affected by the attack due to the fall in assets. When you know all the financial investment involved in running a mining pool, it is clear that the action will be counterproductive. And even if such a pool were formed, the alert should be automatically launched by the rest of the network about the potential danger it would represent.
While it’s true that Bitcoin’s Proof of Work system seems conducive to executing a 51% attack, its implementation is far from straightforward. Indeed, more active computing and hashing power increases security against this attack in the case of bitcoin. However, small communities – Bitcoin Gold (BTG) case– are more vulnerable and make the attack possible.
In the crypto-bath since 2017, both much too late but quite early from the point of view of my future grandchildren 🙂
Writer, Amateur Trader, Ethereum and NEO-compatible