News - 50 percent of Bitcoin Lightning Nodes run on Amazon and Google
By
Bitcoin stands for decentralization. But what about the Lightning network and its nodes? About 50 percent, meanwhile, run on Amazon and Google.
Bitcoin's holy grail: decentralization. Hardly any other network connects more with this story. Bitcoin aims to use decentralization to solve exactly the problems created by the traditional financial system. To be scalable as a global network, there is Bitcoin Lightning: a supposed solution to the blockchain trilemma. But where Lightning provides fast transactions, decentralization stops.
According to data from Mempool Space, about half of all Bitcoin Lightning Network (LN) nodes run on centralized cloud service providers. Amazon Web Service (AWS) hosts about 28 percent and Google Cloud more than 20 percent.
Unlike using a local node, using centralized solutions provides availability and reliability, at least partially. In the past, there have been outages at Amazon or Google which creates a serious problem for Bitcoin security. The potential dangers are:
Centralization of control: Central hosting providers such as Amazon and Google can indirectly exert significant control over the Lightning network by censoring node operators, for example. This contradicts the decentralized nature of Bitcoin and Lightning, where control should be spread across many participants. If one party has too much control, there is a risk of abuse, censorship or manipulation. A wound that hosting provider Hetzner has already exposed with Ethereum.
Trust issue: Transactions can be conducted without the need for a trusted intermediary. This is the philosophy of decentralization in Bitcoin and Lightning. If many of the nodes go through a central provider, users must trust that provider, at least in part.
Single point of failure: The danger of centralized providers is the single point of failure. For example, if Amazon or Google are hacked or heavily regulated, some of Lightning's decentralization is lost.
To avoid increasing centralization, it is important that Lightning nodes be managed by a large number of independent participants. After all, the remaining 50 percent of node operators are distributed among various hosting providers. Still, a better decentralized distribution would be better, keeping in mind the Bitcoin philosophy.
While the lightning network must process many transactions as quickly as possible, the main network limps along like an old donkey. The blockchain processes only seven transfers per second, and that's just as well. After all, Bitcoin says, "A feature, not a bug." It is precisely because the network is so slow that it is so secure, with Proof of Work backing it up.
Miners compete with each other to find the next block. Those who provide more computing power which is also called hashrate have better chances. It used to be that anyone could afford it: Bitcoin mining was possible on almost any commercially available computer. Today, Bitcoin mining is a billion-dollar business with large monopolies. About half of the total hashrate is provided by two mining companies: Foundry USA and AntPool.
The distribution among nodes also leaves much to be desired. About 31 percent of all Bitcoin nodes run in the US. Germany also contributes to centrality with nearly 15 percent. Why even location plays a role in decentralization is shown by the example of China. In the summer of 2021, the Chinese government declared bitcoin mining illegal. Within a very short time, the hashrate plummeted. Before the ban, 70 exahashes per second were coming out of the Asian republic, nearly half of the total computing power in the network. As for the Lightning nodes: Amazon accounts for just under five percent, Google disappears behind Deutsche Telekom AG by two percent.
The promise of blockchain is clear: it is transparent, immutable and above all - decentralized. Meanwhile, however, decentralization is often nothing more than a marketing term. Actual decentralization is difficult to recognize. Until now, people often used the Nakamoto coefficient, which determines the number of nodes needed to disrupt a blockchain network. But other factors also play a role, such as geographic location and the hardware and software infrastructure used.