Partnered With:

Ethereum Foundation: ETH will move to PoS in the coming months, reducing energy consumption by at least 99.95%

Read this article in 8 Minutes

Ethereum's energy consumption could be reduced by at least 99.95% after the merger.

A country's worth of power, no more

Original source: Carl Beekhuizen, Ethereum Foundation

Last night's porridge, Babbitt

Long story short: Ethereum could reduce its energy consumption by at least 99.95% once the merger is completed.

Ethereum will complete the transition to a proof-of-stake (POS) consensus mechanism in the next few months, which will lead to countless improvements that have been theorized. But now that the Beacon chain has been up and running for a few months, we can actually drill down into the numbers. One area we are excited to explore involves new energy use estimates, because we will end the process of spending the value of a country's energy consumption on consensus.

As of now, there aren't any specific statistics on energy consumption (or even what hardware is used), so here's a rough estimate of how much energy Ethereum will consume in the future.

Since many people are running multiple validators, I decided to use the number of independent addresses that can be deposited as a proxy for how many servers there are today. Many pledgers can use multiple ETH 1.0 addresses, but this largely cancels out those redundant Settings.

At the time of this writing, there are 140592 validators from 16,405 separate addresses. Obviously, this is due to the bias created by exchanges and staking pledge services, so removing them would result in 87,897 validators being assumed to be mortgaged at home. As a sound check, this means that the average home pledger runs 5.4 validators, which seems to me to be a reasonable estimate.

Energy requirements

How much energy does it take to run one beacon node (BN), 5.4 validator clients (VC), and one Ethereum 1.0 full node? Based on my personal setup, it's about 15 watts. Joe Clapis (Rocket Pool developer) recently ran 10 validator clients (VCs), 1 Nimbus Beacon node (BN), and 1 Geth full node of a 10Ah USB battery pack for 10 hours, This means that this setting averages 5 watts. The average investor is unlikely to run such an optimization, so we use 100 watts as a reference number.

Multiplier that by the previous 87,000 validators, and that translates to a home pledge consumption of about 1.64 megawatts. The energy consumption of the managed pledge is expected to be slightly higher, as they run thousands of redundant and backup validator clients.

To simplify the calculation, we also assume they use 100 watts per 5.5 validators. Based on the infrastructure teams I've talked to, this is a rough high estimate. The real answer is about 50 times less (if you're a pledge hosting team and each validator consumes more than 5 watts of power, I'm sure I can help you).

So, overall, the proof-of-stake (POS) Ethereum network consumes about 2.62 megawatts of power. That's not the size of a country, or a province or even a city, but about the size of a small town (about 2,100 American homes).

For reference, the current Proof of Work (PoW) Ethereum network consumes as much energy as a medium-sized country, but this is actually necessary to keep the PoW chain secure. As the name suggests, POW reached consensus on the basis of which fork did the most "work" in this regard. There are two ways to increase the "work" completion rate, one is to make mining hardware more efficient, and the other is to use more hardware at once. To prevent a successful attack on the blockchain, miners must "work" faster than the attackers. Since an attacker is likely to have similar hardware, a miner must keep a lot of efficient hardware running in case an attacker digs it up, all of which consume a lot of power.

Under the POW consensus mechanism, the ETH price is positively correlated with computing power. So as prices rise, in equilibrium, the network consumes more electricity. Under the POS consensus mechanism, when the price of ETH increases, the security of the network will also increase (the value of ETH is higher), but the demand for energy remains the same.

A few more

Digital economists estimate that Ethereum miners currently consume 44.49 terawatt-hours of electricity per year, which means that POS is about 2,000 times more energy efficient based on the above conservative estimate, reflecting at least a 99.95% reduction in total energy use.

If the power consumption per transaction is higher than your speed, it's about 35Wh/tx (about 60K gas/tx on average) or about 20 minutes of TV. By comparison, ethereum POW uses 2.8 days' worth of energy per transaction, and bitcoin uses 38 days' worth of energy per transaction.

Looking to the future

While Ethereum is still using the PoW consensus mechanism for now, it won't be for long. In The past few weeks, we have seen The emergence of The first test networks for Merges (note: The Merge is The name Ethereum used when switching from PoW to PoS). Several teams of engineers are working overtime to ensure that the merger happens as quickly as possible without compromising safety.

Scaling solutions, such as rollup and sharding, will help further reduce the energy consumed per transaction by leveraging economies of scale.

The days of the ethereum network being super power-hungry are numbered, and I hope so are the rest of the industry.