Skip to content
Merged
Changes from 1 commit
Commits
Show all changes
25 commits
Select commit Hold shift + click to select a range
e97239c
add energy consumption page
jmcook1186 Aug 18, 2021
1bfdfc5
update index.md
jmcook1186 Aug 18, 2021
2e811b1
Merge branch 'ethereum:dev' into dev
jmcook1186 Aug 18, 2021
9c22ab9
Merge branch 'ethereum:dev' into dev
jmcook1186 Aug 19, 2021
1bfa4a0
Merge branch 'ethereum:dev' into dev
jmcook1186 Sep 7, 2021
8258e28
Merge branch 'ethereum:dev' into dev
jmcook1186 Sep 8, 2021
840e228
amended according to PR review comments
jmcook1186 Sep 15, 2021
d63b5bc
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
2079d73
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
414459c
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
297c29d
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
84c8b1e
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
efb04d3
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
58ab148
Update src/content/energy_consumption/index.md
minimalsm Sep 17, 2021
e4f0abe
correct error in no. beacon validators
jmcook1186 Sep 17, 2021
7000f9b
Merge branch 'dev' of https://github.com/jmcook1186/ethereum-org-webs…
jmcook1186 Sep 17, 2021
73e352d
Apply suggestions from code review
minimalsm Nov 2, 2021
ea0e63b
Update index.md
minimalsm Nov 2, 2021
68e4ab7
Merge branch 'ethereum:dev' into dev
jmcook1186 Nov 3, 2021
23e7da7
Merge branch 'ethereum:dev' into dev
jmcook1186 Nov 9, 2021
23a75fc
Updates based on suggestions
minimalsm Nov 9, 2021
7a61402
add refs as per PR review comments
jmcook1186 Nov 9, 2021
2618438
add refs
jmcook1186 Nov 10, 2021
eaa529f
Apply suggestions from code review
minimalsm Dec 9, 2021
fee0828
Update src/content/energy_consumption/index.md
minimalsm Dec 9, 2021
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Updates based on suggestions
  • Loading branch information
minimalsm committed Nov 9, 2021
commit 23a75fc78b96832fde3a84882c74afffc329195b
32 changes: 23 additions & 9 deletions src/content/energy_consumption/index.md
Original file line number Diff line number Diff line change
Expand Up @@ -7,7 +7,7 @@ sidebar: true

Ethereum's current energy expenditure is too high and unsustainable. Resolving energy expenditure concerns without sacrificing security and decentralization has been a significant technical challenge for Ethereum's development. Let's explore why building Ethereum has had a high environmental impact and how upcoming network upgrades will dramatically change this.
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Ethereum's current energy expenditure is too high and unsustainable.

Agree with the sentiment here but I'm not sure leading this page with a subjective statement is the right approach. It's "too high" - according to who? It's "unsustainable" - is it? Smells like someones opinion.

I'd suggest focusing more on factual statements. e.g. "Ethereum's current energy expenditure is significant - comparable to the power consumption of a small country". We could potentially link out to sources here (e.g. https://digiconomist.net/ethereum-energy-consumption/ though I'm not sure this is an authoritative source).

"Overall, Ethereum represents around 0.1% of global electricity" might be another useful stat for context:
https://kylemcdonald.github.io/ethereum-emissions/

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Agree.

Maybe a more high-level critique about the intro and the order, I wish the article started with explaining the reduced energy impact under PoS and then went lightly into the temporary impact we have under PoW (sort of how Carl's article did). Our current situation is temporary after all ... it's kind of like we're burying the lead by putting the PoS information at the bottom.

I think a quick remedy could be putting a tl;dr that links to Carl's article at the top, something that says like:

TL;DR: Ethereum will use at least ~99.95% less energy with the transition to Proof of Stake. Read more [here](https://blog.ethereum.org/2021/05/18/country-power-no-more/).

Would be cool if that had it's own box to stand our, sort of like on the history page:

Screen Shot 2021-12-08 at 9 45 53 PM


## Energy secures the network {energy-secures-the-network}
## Energy secures the network {#energy-secures-the-network}

Transactions on the Ethereum blockchain are validated by [miners](/developers/docs/consensus-mechanisms/pow/mining). Miners bundle together transactions into ordered blocks and add them to the Ethereum blockchain. The new blocks get broadcast to all the other node operators who run the transactions independently and verify that they are valid. Any dishonesty shows up as an inconsistency between different nodes. Honest blocks are added to the blockchain and become an immutable part of history.

Expand All @@ -17,13 +17,13 @@ Ethereum has used proof-of-work since genesis. Migrating off of proof-of-work ha

## Proof-of-work energy expenditure {#proof-of-work}

Proof-of-work is a robust way to secure against dishonest changes to the blockchain, but it is problematic for several reasons. Since the right to mine a block requires solving a computational puzzle, miners can increase their odds of success by investing in more powerful hardware. These incentives cause an arms race with miners acquiring increasingly power-hungry mining equipment. Ethereum's proof-of-work protocol currently consumes as much energy as a medium-sized country.
Proof-of-work is a robust way to secure against dishonest changes to the blockchain, but it is problematic for several reasons. Since the right to mine a block requires solving a computational puzzle, miners can increase their odds of success by investing in more powerful hardware. These incentives cause an arms race with miners acquiring increasingly power-hungry mining equipment. Ethereum's proof-of-work protocol currently consumes as much energy as a medium-sized country<sup>[^1]</sup>.

## Proof-of-stake {#proof-of-stake}

A greener future for Ethereum is already being built in the form of a **proof-of-stake (PoS)** chain. Under proof-of-stake, arbitrary puzzle-solving is unnecessary. Removing puzzle-solving drastically reduces the energy expenditure required to secure the network. Miners get replaced by validators who perform the same function except that instead of expending their assets up-front in the form of computational work, they stake ETH as collateral against dishonest behavior. If the validator's node is non-responsive or a fraudulent block gets submitted to the chain, the staked assets can be "slashed", strongly incentivizing honesty and securing the network.
A greener future for Ethereum is already being built in the form of a **proof-of-stake (PoS)** chain. Under proof-of-stake, arbitrary puzzle-solving is unnecessary. Removing puzzle-solving drastically reduces the energy expenditure required to secure the network. Miners get replaced by validators who perform the same function except that instead of expending their assets up-front in the form of computational work, they stake ETH as collateral against dishonest behavior. If the validator submits a provably fradulent block to the chain, the staked assets will be "slashed", strongly incentivizing honesty and securing the network.

Similarly to proof-of-work, to maintain a fraudulent blockchain, a validator would require 51% of the total ETH staked in the network. However, unlike proof-of-work, consensus is not based on the longest chain, but a mechanism known as ["Casper"](https://arxiv.org/abs/1710.09437). Migrating from proof-of-work to proof-of-stake eliminates the need to expend energy on arbitrary computations.
Similarly to proof-of-work, to maintain a fraudulent blockchain, a validator would require 51% of the total ETH staked in the network. However, unlike on proof-of-work, where the potential loss of a failed attack is only the cost of generating the hash power needed to mine, on proof-of-stake, the possible loss of an attack is the entire amount of ETH used as collateral. This disincentive structure allows for network security with proof-of-stake while eliminating the need to expend energy on arbitrary computations.

## The merge {#the-merge}

Expand All @@ -35,10 +35,12 @@ As well as building confidence in the proof-of-stake mechanism, the Beacon Chain

![image](energy_use_per_transaction.png)

We can use this data to compare Ethereum to a global service like Visa. 100,000 Visa transactions uses 149kWh of energy<sup>[^2]</sup>. Assuming sharding has been implemented, Ethereum's current transaction rate (15 transactions per second) will be increased by at least 64x (the number of shards), not accounting for additional optimization from rollups. A realistic estimate for post-merge, sharded Ethereum with rollups is [25000 - 100000](https://twitter.com/VitalikButerin/status/1312905884549300224?s=20) transactions per second. We can use this information to estimate a maximum and minimum energy expenditure per 100,000 transactions.
<p style="text-align: center;"><small><i>Estimates based on May 2021 data</i></small></p>

- 25000 transactions per second.
- `100,000 / 25000 = 4` seconds to process 100,000 transactions.
Let's compare these numbers to a service such as Visa. 100,000 Visa transactions uses 149kWh of energy<sup>[^2]</sup>. Assuming sharding has been implemented, Ethereum's current transaction rate (15 transactions per second) will be increased by at least 64x (the number of shards), not accounting for additional optimization from rollups. A realistic estimate for post-merge, sharded Ethereum with rollups is [25,000 - 100,000](https://twitter.com/VitalikButerin/status/1312905884549300224?s=20) transactions per second. We can use this information to estimate a maximum and minimum energy expenditure per 100,000 transactions.

- 25,000 transactions per second.
- `100,000 / 25,000 = 4` seconds to process 100,000 transactions.

We can also estimate Ethereum's energy expenditure per second, making a conservative estimate that 10,000 active validators are securing the network (there are over 180,000 validators on the Beacon Chain at the moment, but many validators can operate on a single node. Currently, there are 3000-4000 individual nodes, so 10,000 is a conservative estimate for post-merge):
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is there a page we can link out to to see these stats in real time?

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@wackerow correct me if I'm wrong but these are rough deductions — I don't think it's possible to accurately measure this?

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Beaconscan has real time info on validator count on the home page https://beaconscan.com/ - will link here. I don't have a citation for the number of nodes though - the 10,000 estimate is a conservative estimate borrowed from the NFT page.


Expand All @@ -51,9 +53,9 @@ This is ~0.4% of the energy used by Visa for the same number of transactions, or

Repeating the calculation with the maximum transactions-per-second yields 0.1667 kWh per second which is about 0.1% of the energy expenditure of Visa, or a reduction of ~894x.

_We’ve provided the basic comparison to Visa to baseline your understanding of post-merge Ethereum energy consumption against a familiar name. However, in practice, it’s not really correct to compare based on number of transactions. Ethereums energy output is time-based. If Ethereum did more or less transactions from one minute to the next, the energy output would stay the same._
_It's not entirely accurate to compare based on number of transactions as Ethereum's energy usage is time-based. The energy usage of Ethereum is the same in 1 minute regardless if it does 1 or 1,000 transactions._

_It’s also important to remember that Ethereum does more than just financial transactions, it’s a platform for applications, so a fairer comparison might be to many companies/industries including Visa, AWS and more!_
_We must also consider that Ethereum isn't limited to simple financial transactions but is also a complete platform built for smart contracts and decentralized applications._
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I feel like we could merge this sentence into the one above, as a set of italicized disclaimers.


## A greener Ethereum {#green-ethereum}

Expand All @@ -74,3 +76,15 @@ While Ethereum's energy consumption has historically been substantial, there has
- [The Beacon Chain](/eth2/beacon-chain)
- [The merge](/eth2/merge/)
- [Sharding](/eth2/beacon-chain/)

### Footnotes and sources {#footnotes-and-sources}

#### 1. Ethereum proof-of-work energy consumption {#fn-1}

[Energy Consumption by Country inc. Ethereum (Annualized TWh)](https://digiconomist.net/ethereum-energy-consumption)

#### 2. Visa energy consumption {#fn-2}

[Bitcoin network average energy consumption per transaction compared to VISA network as of 2020, Statista](https://www.statista.com/statistics/881541/bitcoin-energy-consumption-transaction-comparison-visa/)

[Visa financials report Q4 2020](https://s1.q4cdn.com/050606653/files/doc_financials/2020/q4/Visa-Inc.-Q4-2020-Operational-Performance-Data.pdf)